two-dimensional self-assembly of amphiphilic block copolymers...
TRANSCRIPT
1
TWO-DIMENSIONAL SELF-ASSEMBLY OF AMPHIPHILIC BLOCK COPOLYMERS AT THE AIR/WATER INTERFACE AND NANOPARTICLES FOR DRUG DETOXIFICATION
APPLICATIONS
By
THOMAS J. JONCHERAY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2006
2
Copyright 2006
by
Thomas J. Joncheray
3
ACKNOWLEDGMENTS
First and foremost, I would like to thank my parents, Dominique Joncheray and Catherine
Stona, my sister, Alice Joncheray, as well as the rest of my family, past and present, for being an
endless source of support in my education and life. In February 2001, I met Emilie Galand in
Bordeaux who came to the University of Florida with me where we spent together almost five
years of graduate school. I could not have achieved this seemingly insurmountable amount of
work without her. We supported each other through these challenging years, and helped each
other through the difficult times.
I would like to acknowledge my research director, Prof. Randolph S. Duran, for his help
and support over the years I spent under his supervision. His experienced advice has been an
integral part of my education and has given me a deeper understanding of what is needed to be
successful as a research chemist. I would like also to thank the other members of my Ph.D.
committee: Prof. Kenneth B. Wagener, Dr. Ronald K. Castellano, Dr. Thomas J. Lyons, and Dr.
Wolfgang M. Sigmund.
I am also very grateful to the many collaborators I have had the chance to interact with
over the years I spent in graduate school. I have had the pleasure of working with Prof. Audebert
from ENS Cachan on the nanocapsule project. He provided very interesting discussions and
ideas, and always made himself available when I needed his help. I also really appreciated the
collaboration on PEO-b-PCL block copolymers I had with Prof. Schubert and particularly with
Dr. Mike Meier from the Eindhoven University of Technology. It was a pleasure to have Mike
and Jutta staying for a few days at the University of Florida, and Emilie and I also enjoyed very
much the time spent in Eindhoven. I express my appreciation to Prof. Gnanou from the
Laboratoire de Chimie des Polymères Organiques in Bordeaux, France, for his input in the PS-b-
4
PtBA, PS-b-PAA, and PB-b-PEO projects, and for helping me in joining the graduate program of
the University of Florida.
At the University of Florida, special thanks go to the polymer floor and the Chemistry
Department staff, most notably Lorraine Williams, Sara Klossner, and Lori Clark for their
patience in answering my various questions. I want to show my gratitude to the three professors
of the Butler Polymer Laboratory, Prof. Duran, Prof. Wagener, and Prof. Reynolds, for putting a
lot of effort in providing the polymer floor members with a superior work environment to
conduct research in polymer chemistry. I want to thank all my co-workers from the Duran group
and from the George and Josephine Butler polymer floor. Special thanks need to go to Rachid
Matmour, who made the time spent in the lab really enjoyable. It was a pleasure working with
him, often in a hilarious atmosphere, on several challenging research projects. Other group
members to whom I owe extra thanks for their help along the way are Dr. Aleksa Jovanovic, Dr.
Jennifer Logan, Jorge Chávez, Sophie Bernard, Brian Dorvel, Rita El-Khouri, Kristina
Denoncourt, and Claire Mathieu.
Finally, I would like to thank all the friends I have had the chance to meet and interact with
inside or outside the Chemistry Department, especially Benoît Lauly, Christophe Grenier, Pierre
Beaujuge, and Changhwan Ko who made my days in Gainesville very enjoyable.
5
TABLE OF CONTENTS page
ACKNOWLEDGMENTS ...............................................................................................................3
LIST OF TABLES...........................................................................................................................8
LIST OF FIGURES .........................................................................................................................9
ABSTRACT...................................................................................................................................15
CHAPTER
1 INTRODUCTION ..................................................................................................................17
1.1 Block Copolymers in the Bulk and in Solution ................................................................17 1.2 Block Copolymers at the A/W Interface ..........................................................................20 1.3 Current Status of Drug Detoxification Therapy ...............................................................23 1.4 Nanoparticle Technology..................................................................................................24 1.5 Microemulsions and Sol-Gel Chemistry ..........................................................................25 1.6 Molecular Imprinting and Miniemulsion Polymerization ................................................28
2 EXPERIMENTAL TECHNIQUES........................................................................................32
2.1 Langmuir Monolayers and Surface Pressure Related Experiments .................................32 2.2 Langmuir-Blodgett Films and Atomic Force Microscopy ...............................................36 2.3 Transmission Electron Microscopy and Quasi-Elastic Light Scattering..........................39 2.4 Cyclic Voltammetry..........................................................................................................41
3 POLYSTYRENE-b-POLY(TERT-BUTYLACRYLATE) AND POLYSTYRENE-b-POLY(ACRYLIC ACID) DENDRIMER-LIKE COPOLYMERS: TWO-DIMENSIONAL SELF-ASSEMBLY AT THE AIR-WATER INTERFACE......................44
3.1 Introduction.......................................................................................................................44 3.2 Results and Discussion .....................................................................................................46
3.2.1 PS-b-PtBA Dendrimer-Like Copolymer ................................................................46 3.2.2 PS-b-PAA Dendrimer-Like Copolymer .................................................................52
3.3 Conclusions.......................................................................................................................58 3.4 Experimental Methods......................................................................................................59
3.4.1 Materials .................................................................................................................59 3.4.2 Langmuir Films ......................................................................................................59 3.4.3 AFM Imaging .........................................................................................................60
4 LANGMUIR AND LANGMUIR-BLODGETT FILMS OF POLY(ETHYLENE OXIDE)-b-POLY(ε-CAPROLACTONE) STAR-SHAPED AND LINEAR BLOCK COPOLYMERS .....................................................................................................................61
6
4.1 Introduction.......................................................................................................................61 4.2 Results and Discussion .....................................................................................................64
4.2.1 PEO Homopolymers...............................................................................................64 4.2.2 PCL Homopolymers...............................................................................................67 4.2.3 Star-Shaped PEO-b-PCL Block Copolymers.........................................................73
4.2.3.1 High MMA region........................................................................................73 4.2.3.2 Intermediate MMA region ...........................................................................77 4.2.3.3 Low MMA region ........................................................................................79 4.2.3.4 AFM imaging ...............................................................................................81
4.2.4 PEO-b-PCL Linear Diblock Copolymers...............................................................83 4.2.4.1 Low surface pressure region ........................................................................85 4.2.4.2 High surface pressure region........................................................................90
4.3 Conclusions.......................................................................................................................95 4.4 Experimental Methods......................................................................................................96
4.4.1 Langmuir Films ......................................................................................................96 4.4.2 AFM Imaging .........................................................................................................97
5 TWO-DIMENSIONAL POLYMERIC NANOMATERIALS THROUGH CROSS-LINKING OF POLYBUTADIENE-b-POLY(ETHYLENE OXIDE) MONOLAYERS AT THE AIR/WATER INTERFACE....................................................................................98
5.1 Introduction.......................................................................................................................98 5.2 Results and Discussion ...................................................................................................100
5.2.1 Hydrosilylated PB Homopolymer ........................................................................100 5.2.1.1 Hydrosilylation reaction.............................................................................100 5.2.1.2 Cross-linking reaction at the A/W interface...............................................103 5.2.1.3 AFM imaging .............................................................................................107
5.2.2 Hydrosilylated PB-b-PEO Three-Arm Stars ........................................................109 5.2.2.1 Hydrosilylation reaction.............................................................................112 5.2.2.2 Cross-linking reaction at the A/W interface...............................................115 5.2.2.3 AFM imaging .............................................................................................119
5.3 Conclusions.....................................................................................................................124 5.4 Experimental Methods....................................................................................................125
5.4.1 Materials and Instrumentation..............................................................................125 5.4.2 Langmuir Films ....................................................................................................126 5.4.3 Hydrosilylation of the PB Homopolymer.............................................................126 5.4.4 Hydrosilylation of the (PB200-b-PEO326)3 Three-Arm Star Block Copolymer....127 5.4.5 A/W Interfacial Cross-Linking.............................................................................128
6 ELECTROCHEMICAL AND SPECTROSCOPIC CHARACTERIZATION OF ORGANIC COMPOUND UPTAKE IN SILICA CORE-SHELL NANOCAPSULES ......129
6.1 Introduction.....................................................................................................................129 6.2 Results and Discussion ...................................................................................................131
6.2.1 Nanocapsule Characterization ..............................................................................133 6.2.2 Uptake Study ........................................................................................................134
6.2.2.1 Optical measurements results.....................................................................136
7
6.2.2.2 Electrochemical experiments .....................................................................138 6.3 Conclusions.....................................................................................................................146 6.4 Experimental Methods....................................................................................................147
6.4.1 Nanocapsule Synthesis ........................................................................................147 6.4.2 Transmission Electron Microscopy......................................................................147 6.4.3 Particle Size Analysis ...........................................................................................148 6.4.4 Spectroscopy Measurements ................................................................................148 6.4.5 Electrochemistry Experiments..............................................................................149
7 TOWARD SPECIFIC DRUG DETOXIFICATION AGENTS: MOLECULARLY IMPRINTED NANOPARTICLES.......................................................................................151
7.1 Introduction.....................................................................................................................151 7.2 Results and Discussion ...................................................................................................152 7.3 Conclusions.....................................................................................................................160 7.4 Experimental Section......................................................................................................161
7.4.1 Materials ...............................................................................................................161 7.4.2 Nanoparticle Synthesis .........................................................................................162 7.4.3 FTIR Spectroscopy...............................................................................................162 7.4.4 Particle Size Analysis ...........................................................................................162 7.4.5 AFM Imaging .......................................................................................................163 7.4.6 Uptake Experiments .............................................................................................163
8 CONCLUSIONS AND PERSPECTIVES ...........................................................................165
LIST OF REFERENCES.............................................................................................................168
BIOGRAPHICAL SKETCH .......................................................................................................184
8
LIST OF TABLES
Table page 4-1 Characteristic values of the star-shaped polymers.............................................................63
4-2 Characteristic values of the linear PEO macroinitiator and of the linear diblock copolymers.........................................................................................................................64
4-3 Collapse pressure values of the PCL homopolymers. .......................................................69
5-1 Number average molecular weights and polydispersity indexes of the three-arm star block copolymers. ............................................................................................................110
7-1 Loading compositions of the miniemulsions. ..................................................................154
9
LIST OF FIGURES
Figure page 1-1 Mean-field predication of the morphologies for conformationally symmetric diblock
melts. ..................................................................................................................................18
1-2 Solution state for amphiphilic diblock copolymers in water for concentrations below and above the CMC. ..........................................................................................................19
1-3 Spherical, rodlike, and vesicular morphologies for PS-b-PAA crew-cut micelles............20
1-4 A schematic illustration showing the components of an amphiphile and the orientation of this amphiphile adopted at an interface.......................................................21
1-5 Transmission electron micrographs of LB films of poly(styrene-b-vinylpyridinium decyl iodide) AB diblock copolymers ...............................................................................23
1-6 Structure of normal and inverse spherical micelles formed in microemulsion systems....26
1-7 Sol-Gel hydrolysis and condensation reactions. ................................................................27
1-8 Sol-Gel technologies and their products ............................................................................28
1-9 Outline of the molecular imprinting strategy.....................................................................29
1-10 Examples of commercially available functional monomers and cross-linkers..................29
1-11 Principal of miniemulsion polymerization.........................................................................31
2-1 The Langmuir Teflon trough .............................................................................................32
2-2 A Wilhelmy plate partially immersed in a water subphase ...............................................33
2-3 Schematic isotherm for small amphiphilic molecules .......................................................35
2-4 LB film transfer onto a hydrophilic mica substrate ...........................................................37
2-5 The AFM tapping mode electronic setup...........................................................................38
2-6 Specimen interactions in electron microscopy ..................................................................39
2-7 Basic shape of the current response for a cyclic voltammetry experiment........................42
3-1 Schematic sketch of the PS-b-PtBA (Dend1) and the PS-b-PAA (Dend2) samples. ........45
3-2 Isotherm of Dend1. (Inset) Static elastic modulus plot versus surface pressure. ..............47
10
3-3 Topographic AFM images of Dend1 LB films transferred at 5, 10, 15, 20, 24 (middle of plateau, MMA = 20,000 Å2), and 40 mN/m..................................................................49
3-4 Compression-expansion hysteresis plot of Dend1 (target MMA = 20,000 Å2).................50
3-5 Surface pressure/time isochoric relaxation plot of Dend1 after compression up to 40 mN/m .................................................................................................................................51
3-6 Compression-expansion hysteresis plot of Dend1 (target pressure = 40 mN/m) ..............52
3-7 Isotherm of PAA250K (pH = 2.5). .......................................................................................53
3-8 Topographic AFM images of PAA250K LB films transferred at 1.5, 3, and 3.5 mN/m. ....54
3-9 Isotherm of Dend2 (pH = 2.5). (Inset) Compressibility plot versus surface pressure. ......55
3-10 Topographic AFM images of Dend2 LB films transferred at 2, 4, 4.5, 5, 5.5, 6, and 8 mN/m. ................................................................................................................................57
3-11 Compression-expansion hysteresis plot of Dend2 (target pressure = 5 mN/m, pH = 2.5). ....................................................................................................................................57
4-1 The star-shaped PEO-b-PCL block copolymers. ...............................................................63
4-2 The linear PEO-b-PCL block copolymers. ........................................................................64
4-3 Isotherms of the PEO homopolymers. (Inset) Same isotherms normalized with respect to the number of ethylene oxide units. ..................................................................66
4-4 Compression-expansion hysteresis plot of the PEO core (target pressure = 2 mN/m)......66
4-5 Isotherms of the PCL homopolymers. (Inset) Same isotherms normalized with respect to the number of ε-caprolactone units. ..................................................................68
4-6 Isotherms of the PCL homopolymers (compression speed = 100 mm/min). ....................70
4-7 Compression-expansion hysteresis plot of PCL1250. .......................................................71
4-8 Topographic AFM images of PCL homopolymers LB films transferred below and above monolayer collapse..................................................................................................72
4-9 Isotherms of the star-shaped PEO-b-PCL copolymers. .....................................................73
4-10 Compression-expansion hysteresis plot of Star#6 (target pressure = 9 mN/m). ...............74
4-11 Compression-expansion hysteresis plot of Star#1 (target pressure = 9 mN/m). ...............74
4-12 Plots of MMA versus number of ε-caprolactone repeat units for different surface pressures from the isotherms of Star#3, Star#4, Star#5, and Star#6..................................76
11
4-13 Isotherms of the PEO core extrapolated and PEO2000 normalized with respect to the number of ethylene oxide units..........................................................................................76
4-14 Compressibility plots of the star-shaped PEO-b-PCL block copolymers versus surface pressure..................................................................................................................77
4-15 Isobaric relaxation plots of Star#6 at 9, 11, and 13 mN/m ................................................78
4-16 Compression-expansion hysteresis plot of Star#6 (target pressure = 15 mN/m) ..............79
4-17 Compression-expansion hysteresis plot of Star#1 (target pressure = 15 mN/m) ..............79
4-18 Proposed conformations modeling the adsorption of the star-shaped block copolymers at the A/W interface versus surface pressure. ................................................80
4-19 Topographic AFM images of the star-shaped PEO-b-PCL copolymers LB films transferred below and above the pseudoplateau ................................................................82
4-20 Isotherms of the PEO-b-PCL linear diblock copolymers. .................................................84
4-21 Compressibility plots of the PEO-b-PCL linear diblock copolymers versus surface pressure. .............................................................................................................................85
4-22 Isotherms of PEO2670 and PCL2000 binary mixtures. (Inset) Corresponding compressibility plots versus surface pressure. ...................................................................86
4-23 Compression-expansion hysteresis plot of the binary mixture with 49 mol % in PEO2670 (target pressure = 9 mN/m). ..............................................................................86
4-24 MMA plots versus mole fraction of PCL2000. Dashed lines: theoretical ideal mixing....87
4-25 Compression-expansion hysteresis plot of PEO60-b-PCL11 (target pressure = 18 mN/m). ...............................................................................................................................89
4-26 Compressibility plots of Figure 4-25 (PEO60-b-PCL11, target pressure = 18 mN/m)........89
4-27 Compression-expansion hysteresis plot of PEO60-b-PCL35 (target pressure = 16 mN/m). ...............................................................................................................................91
4-28 Compressibility plots of Figure 4-27 (PEO60-b-PCL35, target pressure = 16 mN/m)........92
4-29 Topographic AFM images of the linear PEO-b-PCL diblock copolymers LB films transferred after crystallization of the PCL segment at the A/W interface (π = 15 mN/m) ................................................................................................................................93
4-30 Proposed conformations modeling the adsorption of the linear PEO-b-PCL diblock copolymers at the A/W interface versus surface pressure .................................................94
5-1 Hydrosilylation of the pendant double bonds of the PB homopolymer...........................101
12
5-2 1H NMR spectrum of the PB homopolymer. ...................................................................101
5-3 1H NMR spectrum of the hydrosilylated PB homopolymer. ...........................................102
5-4 FTIR spectra of the PB homopolymer before and after hydrosilylation. ........................102
5-5 Cross-linking reaction involving hydrolysis and condensation of the triethoxysilane groups...............................................................................................................................104
5-6 Surface pressure-Mean Molecular Area isotherms of the hydrosilylated PB carried out after different reaction times (subphase pH = 3.0). ...................................................104
5-7 Static elastic modulus-surface pressure curves of the hydrosilylated PB homopolymer at different reaction times (subphase pH = 3.0). ......................................105
5-8 MMA-time isobars of the hydrosilylated PB for various subphase pH values (π = 10 mN/m). .............................................................................................................................106
5-9 Removal of the cross-linked hydrosilylated PB from the water surface .........................107
5-10 AFM topographic images of the LB films transferred onto mica substrates at π = 10 mN/m. ..............................................................................................................................109
5-11 Surface Pressure-MMA isotherms for the (PB200-b-PEOn)3 three-arm star block copolymers.......................................................................................................................111
5-12 Isotherm of (PB76-b-PEO444)4 depicting how Apancake, Ao, and ∆Apseudoplateau are determined........................................................................................................................111
5-13 Linear dependence of ∆Apseudoplateau on the total number of ethylene oxide units. ...........112
5-14 Hydrosilylation of the pendant double bonds of the (PB-b-PEO)3 three-arm star block copolymers. ............................................................................................................113
5-15 1H NMR spectra of the (PB200-b-PEO326)3 star block copolymer and the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer. ................114
5-16 FTIR spectra of the (PB200-b-PEO326)3 star block copolymer and the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer. ........................................114
5-17 Surface Pressure-MMA isotherms of the (PB200-b-PEO326)3 star block copolymer and of the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer before and after cross-linking. .........................................................................................116
5-18 Surface Pressure-MMA isotherms and compressibility-MMA curves of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer at various reaction times (subphase pH = 3.0). ........................................................................................................117
13
5-19 Isobars of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer for various subphase pH values (π = 5 mN/m)......................................................................118
5-20 Removal of the cross-linked (PB(Si(OEt)3)-b-PEO)3 three-arm star copolymer from the Langmuir trough surface............................................................................................119
5-21 AFM topographic images of the (PB(Si(OEt)3)-b-PEO)3 star block copolymer LB films. ................................................................................................................................122
5-22 Cross-section analysis of Figures 5-21D and 5-21H, PEO pore size versus surface pressure plot, and PSD plots of Figures 5-21C, 5-21D, 5-21G, and 5-21H. ...................122
5-23 AFM topographic images and corresponding cross-sections of the (PB(Si(OEt)3)-b-PEO)3 star block copolymer LB films cross-linked at 9 mN/m (pH = 3.0, t = 10 h) and transferred at 9 and 2 mN/m......................................................................................123
5-24 Surface pressure-MMA isotherms of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer cross-linked at 5 and 20 mN/m (pH = 3.0, t = 10 h). ...........................124
6-1 Oil-filled silica nanocapsule synthesis through initial hydrophobic core formation followed by hydrophilic silica shell formation after TMOS addition..............................132
6-2 Description of the nanocapsule samples prepared using 0.07, 0.28, 0.44, and 0.88 wt % TMOS. .........................................................................................................................132
6-3 DLS results for the microemulsion immediately after preparation and the same solution after TMOS addition (0.07 wt %) and dialysis. .................................................133
6-4 TEM micrographs of the 0.07 wt % TMOS nanocapsules and of the 0.88 wt % TMOS nanocapsules. .......................................................................................................134
6-5 Chemical structures of ferrocene methanol, ferrocene dimethanol, and Nile Red. .........135
6-6 UV-vis absorption spectra of iodine in water solution, in nanocapsule solution, in Tween-80 aqueous solution, and in ethyl butyrate solution. ...........................................136
6-7 Nile Red emission spectra in ethyl butyrate solution, in nanocapsule solution, in Tween-80 aqueous solution, in crushed Xerogel dispersion in acidic water, on silica gel, and in acidic water solution. .....................................................................................137
6-8 Typical cyclic voltammogram of ferrocene methanol in water. ......................................138
6-9 Uptake of ferrocene methanol versus time in 0.07, 0.28, 0.44, and 0.88 wt % TMOS nanocapsule solutions. .....................................................................................................141
6-10 Plot of normalized aqueous concentration of ferrocene methanol after uptake in 0.07, 0.28, 0.44, and 0.88 wt % TMOS nanocapsule solutions ................................................142
14
6-11 Uptake of ferrocene dimethanol versus time in 0.07, 0.28, 0.44, and 0.88 wt % TMOS nanocapsule solutions ..........................................................................................144
6-12 Uptake of ferrocene methanol versus time in 0, 2, 4, 6, and 8 wt % Tween-80 aqueous solutions .............................................................................................................145
6-13 Uptake of ferrocene dimethanol versus time in 0, 2, 4, 6, and 8 wt % Tween-80 aqueous solutions .............................................................................................................146
7-1 Chemical structures of amitriptyline and bupivacaine.....................................................152
7-2 The molecular imprinting strategy in miniemulsion polymerization. .............................154
7-3 IR absorbance spectra of EGDMA, MIP1, and MIP3.......................................................155
7-4 DLS size distribution of MIP6..........................................................................................156
7-5 Tapping mode topographical AFM images and cross-section analysis of MIP6. ............157
7-6 Uptake of amitriptyline by the non-molecularly imprinted nanoparticles MIP1, MIP2, and MIP3 ..........................................................................................................................158
7-7 Uptake of amitriptyline by the nanoparticles molecularly imprinted with amitriptyline: MIP4, MIP5, and MIP6...............................................................................159
7-8 Uptake of bupivacaine by the nanoparticles molecularly imprinted with amitriptyline: MIP4, MIP5, and MIP6...............................................................................160
15
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
TWO-DIMENSIONAL SELF-ASSEMBLY OF AMPHIPHILIC BLOCK COPOLYMERS AT THE AIR/WATER INTERFACE AND NANOPARTICLES FOR DRUG DETOXIFICATION
APPLICATIONS
By
Thomas J. Joncheray
December 2006
Chair: Randolph S. Duran Major Department: Chemistry
The two-dimensional self-assembly at the air/water (A/W) interface of various block
copolymers (dendrimer-like polystyrene-b-poly(tert-butylacrylate) (PS-b-PtBA) and
polystyrene-b-poly(acrylic acid) (PS-b-PAA), linear and five-arm star poly(ethylene oxide)-b-
poly(ε-caprolactone) (PEO-b-PCL), and three-arm star triethoxysilane-functionalized
polybutadiene-b-poly(ethylene oxide) (PB(Si(OEt)3)-b-PEO)) was investigated through surface
pressure measurements (isotherms, isobars, isochores, and compression-expansion hysteresis
experiments) and atomic force microscopy (AFM) imaging. The PS-b-PtBA and the PS-b-PAA
samples formed well-defined circular surface micelles at low surface pressures with low
aggregation numbers (~ 3-5) compared to linear analogues before collapse of the PtBA chains
and aqueous dissolution of the PAA segments take place around 24 and 5 mN/m, respectively.
The linear PEO-b-PCL samples exhibited three phase transitions at 6.5, 10.5, and 13.5 mN/m
corresponding respectively to PEO aqueous dissolution, PEO brush formation, and PCL
crystallization. The two PEO phase transitions were not observed for the star-shaped PEO-b-PCL
samples because of the negligible surface activity of the star-shaped PEO core compared to its
linear analogue. The PB(Si(OEt)3)-b-PEO sample was cross-linked at the A/W interface by self-
16
condensation of the pendant triethoxysilane groups under acidic conditions, which resulted in the
formation of a two-dimensional PB network containing PEO pores with controllable sizes.
With a view toward drug detoxification therapy, the encapsulation abilities of oil core-
silica shell nanocapsules and molecularly imprinted nanoparticles were also investigated by
electrochemical (cyclic voltammetry) and optical (fluorescence and UV-vis spectroscopies)
techniques. The core-shell nanocapsules were shown to efficiently remove large amounts of
organic molecules present in aqueous solutions, with the silica shell acting analogously to a
chromatographing layer. The molecularly imprinted nanoparticles were prepared by the non-
covalent approach and by miniemulsion polymerization. Binding studies on the molecularly
imprinted nanoparticles in aqueous solutions under physiological pH conditions indicated that, in
the absence of specific imprinting, the uptake of toxic drugs was mainly driven by non-specific
hydrophobic interactions. As demonstrated with the use of the antidepressant amitriptyline, in the
presence of specific imprinting the uptake significantly increased as the amount of specific
binding sites was increased.
17
CHAPTER 1 INTRODUCTION
This dissertation aims at summarizing the work realized on two different research domains
of polymer chemistry: the air/water (A/W) interfacial behavior of block copolymers and the
synthesis of nanoparticles for drug detoxification applications. As presented in Chapters 3, 4, and
5, the first three research projects are related to the self-assembly of amphiphilic block
copolymers at the A/W interface, whereas Chapters 6 and 7 describe the investigations carried
out on the possibility for 2 different types of nanoparticulate systems to be used as drug
detoxification agents. While Chapter 2 briefly describes the principal experimental techniques
mentioned in the subsequent chapters, this first introductory chapter serves as literature
background for the different research projects and also defines the key concepts used throughout
this dissertation.
1.1 Block Copolymers in the Bulk and in Solution
“Block copolymer” is a general term used to define a macromolecule composed of
different polymer chains. The field of block copolymers has attracted a lot of interest in the past
thirty years because the eventual phase separation between immiscible blocks in various
environments, such as in the bulk, often leads to well-defined self-assembled structures with
unique morphologies with characteristic sizes ranging between a few nanometers up to hundreds
of nanometers.1,2 Moreover, the recent emergence of controlled polymerization techniques such
as living anionic polymerization,3 ATRP (atom transfer radical polymerization),4 or RAFT
(reversible addition fragmentation chain transfer)5 has allowed access to a variety of
compositions and architectures (star,6 mikto-arm,7 cyclic8. . .), which significantly increases the
diversity of peculiar and regular patterns obtainable resulting from their self-assembly. A lot of
research still has to be done to better understand the relationships between block copolymer
18
architecture and self-assembly. Therefore, because of their simple architecture, linear diblock
copolymers are still currently the best-known class of block copolymers. Several theoretical
models have been proposed to describe the behavior of block copolymers such as for instance the
self-consistent field theory (SCFT)9 or the mean-field theory (MFT)10 where the phase behavior
is dictated by the Flory-Huggins segment-segment interaction parameter, the degree of
polymerization, and the composition. As an example, if the two A and B blocks of a linear AB
diblock copolymer are immiscible, they can adopt in the bulk, as shown in Figure 1-1, various
microphase morphologies such as spheres (S and S’), cylinders (C and C’), double gyroids (G
and G’), or lamellae (L).1c
Figure 1-1. Mean-field predication of the morphologies for conformationally symmetric diblock
melts. Phases are labeled as: S (spheres), C (cylinders), G (double gyroids), L (lamellae). fA is the volume fraction.1c
When block copolymers are dissolved in a selective solvent, the chains can aggregate to
reversibly form well-defined micelles above the so-called critical micellar concentration (CMC).
For concentrations lower than the CMC, block copolymer molecules are unassociated as
illustrated in Figure 1-2 for amphiphilic diblock copolymers in water aggregating into spherical
micelles.
19
Figure 1-2. Solution state for amphiphilic diblock copolymers in water for concentrations below
and above the CMC.
The critical micellar concentrations of block copolymers are usually very low compared to
low molecular weight surfactants, and therefore the micelles formed have great potential in drug
delivery when used as nanocontainers since they hardly dissociate in the blood stream, even
under highly dilute conditions.1a Depending on the architecture, composition, concentration, or
solvent, block copolymers can aggregate into a variety of micellar structures. Figure 1-3 shows
as an example some peculiar morphologies obtained by Eisenberg and co-workers for “crew-cut”
micelles of linear polystyrene-b-poly(acrylic acid) (PS-b-PAA) diblock copolymers in aqueous
solutions.11 Transitions from spheres to rod to vesicles were observed as the length of the PAA
segment was decreased. In semi-dilute or concentrated solutions, gelation normally occurs, and
block copolymer micelles organize into a nanostructure-ordered lyotropic liquid crystal phase.12
20
Figure 1-3. Spherical (a), rodlike (b), and vesicular (c) morphologies for PS-b-PAA crew-cut
micelles.11
1.2 Block Copolymers at the A/W Interface
The behavior of block copolymers at interfaces is also of great interest since other
parameters such as the surface energies as well as film thickness can strongly influence the
microphase separation.1c Confining polymeric chains in a layer thinner than their natural length
scale (the radius of gyration) considerably alters their conformation and the resulting physical
properties compared to the bulk properties.13 Among the variety of surfaces and interfaces
available, the A/W interface has attracted a lot of attention because it allows the easy formation
of two-dimensional polymeric monolayers (Langmuir monolayers), providing that the block
copolymers studied are surface active by having sufficiently polar functional groups to adsorb at
the interface (without being too much water soluble to avoid their irreversible dissolution in the
aqueous subphase). Similarly as for low molecular weight surfactants, surface active block
copolymers self-assemble at the A/W interface to reduce the surface tension (internal pressure
caused by the attraction of molecules below the surface for those at the surface) with the
hydrophilic segments immersed into the water and the hydrophobic segments desorbed in the air
as illustrated in Figure 1-4 for a low molecular weight fatty acid amphiphile.14
21
Figure 1-4. A schematic illustration showing the components of an amphiphile and the
orientation of this amphiphile adopted at an interface.14
As previously shown, one of the great advantages of the A/W interface is that it is possible
and fairly easy to accurately control and adjust the way the chains of the block copolymers self-
assemble simply by varying their surface concentration (amount present at the interface) in
addition to the other usual parameters mentioned before for the bulk or for solutions. This often
leads to peculiar arrangements of the polymer chains with the formation of surface micelles,
which significantly differs from what is commonly observed for low molecular weight
amphiphiles. Control of surface density in the Langmuir films and transfer of the block
copolymer monolayers onto solid substrates for further analysis (Langmuir-Blodgett films) are
experimental procedures commonly done with the use of a Langmuir trough as extensively
described later in Chapter 2. A wide range of experimental techniques can be used for
morphology investigation in Langmuir and Langmuir-Blodgett (LB) monolayers including
neutron and X-ray reflectivity, surface pressure and potential measurements, Brewster angle
microscopy (BAM), atomic force microscopy (AFM), transmission electron microscopy (TEM),
and ellipsometry.15-20 Examples of block copolymers with various architectures previously
22
studied at the A/W interface include poly(ethylene oxide)-b-poly(propylene oxide) (PEO-b-
PPO), polystyrene-b-poly(ethylene oxide) (PS-b-PEO), polybutadiene-b-poly(ethylene oxide)
(PB-b-PEO), and polystyrene-b-polyacrylate.21-24 Block copolymers based on PS and decylated
poly(4-vinylpyridine) (decylated P4VP) have also been extensively investigated at the A/W
interface by Eisenberg and co-workers, and some of their results are presented here to illustrate
the concept of surface micelle formation. Surface micelle formation results from the presence at
the A/W interface of immiscible blocks that phase separate because of sufficiently different
polarities. In the case of PS and decylated P4VP-based block copolymers, only the hydrophilic
decylated P4VP segments adsorb at the A/W interface with the hydrophobic PS segments
desorbing and aggregating above the interface. This results in the formation of well-defined
surface micelles with architectures evolving from circular to rod-like to planar as the PS % in the
diblock copolymers is increased (Figure 1-5).25 Prospective applications of such well-defined
patterns with feature sizes typically in the nanometer scale order include lithographic masks,
photonic materials, and nanopatterned substrates for microelectronics.26-28
The results presented in Chapters 3, 4, and 5 of this dissertation are all related to the self-
assembly of block copolymers at the A/W interface. While the work presented in Chapters 3 (on
polystyrene-b-poly(tert-butylacrylate) and polystyrene-b-poly(acrylic acid) dendrimer-like block
copolymers) and 4 (on linear and star-shaped poly(ethylene oxide)-b-poly(ε-caprolactone) block
copolymers) describes the interfacial aggregation of several block copolymers on a fundamental
level, the work presented in Chapter 5 (on star-shaped poly(ethylene oxide)-b-hydrosilylated
polybutadiene block copolymers and done in collaboration with Rachid Matmour, graduate
student in the Duran group at the University of Florida) focuses primarily on the use of a post
23
monolayer formation cross-linking method using Sol-Gel chemistry (Section 1-5) for the
synthesis of novel two-dimensional cross-linked nanomaterials with well-defined morphologies.
Figure 1-5. Transmission electron micrographs of LB films of poly(styrene-b-vinylpyridinium
decyl iodide) AB diblock copolymers (surface pressure = 2 mN/m) (a) P(S260-b-VP120/C10I); % PS = 68.4; (b) P(S260-b-VP71/C10I); % PS = 78.5; (C) P(S260-b-VP29/C10I) % PS = 90 .0; (d) P(S480-b-VP34C10I) % PS = 93.4; (e) P(S180-b-VP11/C10I) % PS = 94.2; (f) P(S480-b-VP13/C10I) % PS = 97.4.25
1.3 Current Status of Drug Detoxification Therapy
Drug toxicity is a major health concern, because more than 3 million insecticide
poisonings are reported every year worldwide with, among them, around 200,000 fatal cases.29
In the United States, a 2000-query showed that drug-related complications lead to the hospi-
talization of more than 300,000 patients per year. Other important issues are related to clinical
mistreatment for instance in the area of anesthesiology where it was recently shown that the
24
intravenous injections of local anesthetics such as bupivacaine are a major cause of cardiac
blocks with a fatal outcome.30,31 Tricyclic antidepressants such as amitriptyline (one of the most
common drugs used in suicides32) are among the most frequently ingested substances in self-
poisoning.33 Common clinical treatments in the case of anesthetic-induced toxicity include rapid
oxygenation, ventilation, seizure control, and cardiovascular support.34 To fight drug overdose,
several methods have been explored such as the intravenous injection of sodium bicarbonate,35
the use of activated carbon,36 the use of lipid infusions,37 or even the use of other drugs such as
amiodarone,38 insulin,39 or propofol.40 All these methods have shown some interesting results but
an efficient and specific antidote scavenging the toxicity of the aforementioned antidepressant
and anesthetic drugs by reducing their bio-availability within the body still has to be designed.
1.4 Nanoparticle Technology
Living organisms are built of cells that are usually a few micrometers in size, and their
organelles are even much smaller, in the sub-micron size range. The recent development of
nanotechnologies that deal with nanometer-sized objects (nanoparticles) for instance in
electronics, cosmetics, or catalysis41,42 also opened the possibility to probe the cellular machinery
without introducing too much interference43 and to understand biological processes on the
nanoscale level.44 Recent developments of nanoparticulate systems include for instance protein
detection,45 tissue engineering,46 cancer therapy,47 and drug delivery.48 Because of the absence of
specific antidotes in the case of drug intoxication as mentioned in the previous paragraph, a
multidisciplinary effort is being made at the University of Florida (Particle Engineering Research
Center and Departments of Chemistry, Material Science, Chemical Engineering, and
Anesthesiology) to synthesize a series of novel nano-sized bioparticles that are able to move
freely in the blood stream even in the smallest capillaries (diameter ~ 5 µm) and that are able to
25
effectively reduce the free blood concentration of toxic drugs. The ideal nanoparticulate system
therefore should have high and fast (within seconds or minutes) encapsulation capacities, should
be biocompatible (non-toxic), biodegradable (slowly enough so the aqueous concentration of the
drug released in the blood stays below toxic levels), and above all, should be specific to the
target drug to avoid side encapsulation of other undesired molecules present in the blood stream.
Oil core-silica shell nanocapsules have been recently synthesized in the Duran group by
combining microemulsion formulation and Sol-Gel chemistry,49 and with a view toward their
application in detoxification media, the work presented in Chapter 6 focused on investigating
their uptake kinetics and capacities of drug-mimicking compounds. In Chapter 7, a different and
novel type of nanoparticles was synthesized and characterized by combining miniemulsion
polymerization and molecular imprinting technology in an attempt to design a nanoparticulate
system with increased selectivity.
1.5 Microemulsions and Sol-Gel Chemistry
A microemulsion is defined as a system comprising oil, water, and surfactants (surface
active agents having both a hydrophilic part and a hydrophobic part) that results in a single-
phase, optically isotropic, and therefore thermodynamically stable liquid solution. This is not the
case for emulsions that are usually only kinetically stable and that coalesce and phase-separate
over time. When water, oil, and surfactants are mixed together to form a microemulsion, the
surfactant molecules rest at the oil-water interface with the hydrophilic groups solubilized into
the aqueous phase and the hydrophobic groups solubilized into the organic phase to avoid the
direct oil-water contact. The sizes of these structures are usually in the range of a few hundred
nanometers, which makes them suitable for the synthesis of nano-sized objects. Depending on
several parameters such as the nature of the oil or of the surfactant and the concentrations or the
relative amounts of oil, surfactant, and water used, it is possible to obtain a variety of internal
26
structures. They can be spherical (Figure 1-6), spheroidal, or cylindrical “rod” or “worm-like”
micelles, and they may exist in hexagonal, cubic, or lamellar phases. Alternatively, a range of
bicontinuous phases may exist.
Figure 1-6. Structure of normal (left) and inverse (right) spherical micelles formed in microemulsion systems.
Microemulsions can have characteristic properties such as ultra low interfacial tension,
large interfacial area and capacity to solubilize both aqueous and oil-soluble compounds.50
Therefore, they can be used in a variety of applications, for example in enhanced oil recovery,51
as fuels,52 as lubricants, cutting oils, and corrosion inhibitors,53 as coatings and textile finishing,54
in detergency,55 in cosmetics,56 in food,57 in pharmaceuticals,58 or in biotechnology.59 As
presented in Chapter 6, an oil-in-water microemulsion system composed of spherical micelles
was designed in an attempt to selectively encapsulate hydrophobic toxic drugs inside the micellar
oil core and to decrease their free concentration in the blood stream. Even though microemulsion
systems are thermodynamically stable, the high dilution taking place after their injection inside
body fluids would lead to rupture and further coalescence of the micellar structures. This
problem can be avoided by building a solid shell surrounding the microemulsion oil droplets that
will not break upon dilution, for instance by the means of Sol-Gel chemistry (hence the name
core-shell nanocapsules).
27
The Sol-Gel process allows the easy synthesis of ceramic and glass materials with high
purities and homogeneities by using preparation techniques different from the traditional process
of fusion of oxides. This process occurs in liquid solutions in the presence of organometallic
precursors such as metal alkoxides (tetramethoxysilane, tetraethoxysilane, Zr(IV)-propoxide, or
Ti(IV)-butoxide), which, through acid- or base-catalyzed hydrolysis and condensation reactions
as shown in Figure 1-7, leads to the formation of a new phase (Sol).60
Figure 1-7. Sol-Gel hydrolysis and condensation reactions.
The Sol is made of solid particles of a diameter of a few hundred nanometers suspended in
a liquid phase. The particles then condense in a new phase (Gel) in which a solid macromolecule
is immersed in a liquid phase (solvent). Drying the Gel by means of low temperature treatments
allows the formation of materials in a wide variety of forms: ultra-fine or spherical shaped
powders, thin film coatings, ceramic fibers, microporous inorganic membranes, monolithic
ceramics and glasses, or extremely porous aerogel materials as illustrated in Figure 1-8.61 Sol-Gel
materials can be used for instance in chemical sensing,62 drug delivery,63 or electrochromics.64
As further explained in Chapter 6, the formation of a rigid shell to impart stability to the
microemulsion droplets resulting in the formation of core-shell nanocapsules was done by
polycondensation of tetramethoxysilane directly on the microemulsion droplet surface previously
functionalized with trimethoxysilane groups. A similar polycondensation method was also used
as presented in details in Chapter 5 to cross-link in two dimensions block copolymers containing
triethoxysilane-functionalized polybutadiene segments at the A/W interface.
28
Figure 1-8. Sol-Gel technologies and their products.61
1.6 Molecular Imprinting and Miniemulsion Polymerization
The concept of molecularly imprinted polymers (MIPs) was first introduced by Wulff
nearly thirty years ago.65 Since then, the methodology has undergone a number of important
developments, and MIPs have been used in a wide range of techniques (in solid-phase extraction,
as biosensors, as catalysts, and as binding assays) and approaches (covalent, non-covalent, metal
coordination).66-72 MIPs result from the polymerization of monomeric units in the presence of a
template molecule. The imprinting strategy investigated in Chapter 7 is shown in Figure 1-9 and
uses the non-covalent imprinting approach first developed by Mosbach.73,74 Functional
monomers are associated with a template via non-covalent interactions. The complex is then
copolymerized with a cross-linking monomer, followed by removal of the template using
extraction procedures. Removal of the template results in cavities whose shape, size,
functionality, and spatial arrangement are complementary to the imprinted molecule. These
29
recognition sites with predetermined selectivity enable MIPs to rebind selectively the original
template from a mixture. In principle, MIPs can be made with selectivity for essentially any of a
diverse range of analyte species, such as drug enantiomers, pesticides, hormones, toxins, short
peptides, and nucleic acids.75-80
Figure 1-9. Outline of the molecular imprinting strategy.
Many parameters can affect the efficiency of the synthetic molecularly imprinted polymers
such as the porogen (solvent), the polymerization technique, and also the relative concentrations
of the functional monomer, cross-linker, template molecule, and initiator. Figure 1-10 shows
examples of classical functional monomers and cross-linkers commercially available for radical
polymerization.81
Functional monomers
Crosslinkers Figure 1-10. Examples of commercially available functional monomers and cross-linkers.81
30
The main motivation to investigate the potential of molecularly imprinted materials for
drug detoxification applications was to harvest their remarkable ability to rebind specifically a
variety of compounds including toxic drugs. As described later in further details (Chapter 7), the
synthesis of nano-sized molecularly imprinted particles was attempted through miniemulsion
radical polymerization, and their toxic drug-rebinding capacities were investigated.
Miniemulsions are produced by the combination of a high shear to break up the emulsion into
submicron monomer droplets and a surfactant/costabilizer system to retard monomer diffusion
from the submicron monomer droplets.82 The costabilizer is used to prevent Ostwald ripening.83
Compared to microemulsions that usually lead to the formation of oil droplets ranging from 10 to
100 nm, the formation of miniemulsions, which requires a lower amount of surfactants, usually
leads to oil droplet diameters ranging from 50 to 500 nm. The droplet surface area in these
systems is very large, and most of the surfactants are adsorbed at the droplet surface. Since little
surfactant is present in the form of micelles in a properly formulated microemulsion, particle
nucleation during polymerization primarily takes place inside the monomer droplets. This leads
to a 1:1 correspondence between the initial monomer droplets and the final polymer
nanoparticles. The principal of miniemulsion polymerization is schematically shown in Figure 1-
11. As described in greater details in Chapter 7, the synthesis of molecularly imprinted
nanoparticles was done radically, but the miniemulsion polymerization process is not limited to
radical polymerizations.84-86 The imprinting takes place inside the monomer droplets, and
extraction of the toxic drug molecules results in porous nanoparticles with recognition sites
complementary in shape to the original toxic drug template.
31
Figure 1-11. Principal of miniemulsion polymerization.
32
CHAPTER 2 EXPERIMENTAL TECHNIQUES
In this chapter, an overview of the principal experimental techniques is presented. The
specific detailed experimental procedures are not described here but are included instead at the
end of each of the following chapters in the Experimental sections.
2.1 Langmuir Monolayers and Surface Pressure Related Experiments
As further detailed in Chapters 3, 4, and 5, investigation of block copolymer self-assembly
at the air/water (A/W) interface in Langmuir monolayers requires the use of a Langmuir trough.
A typical set-up is shown in Figure 2-1. It provides a very simple method to control monolayer
surface density thanks to the movable barriers controlled by a DC motor. The trough is
composed of an inert material such as for instance Teflon (polytetrafluoroethylene, PTFE) that is
resistant to water absorption.
Figure 2-1. The Langmuir Teflon trough.
To form a Langmuir monolayer, the block copolymer of interest is first dissolved in a
solvent (C ≈ 1mg/mL) which is then spread dropwise on the water surface. The solvent must be
hydrophobic to prevent its dissolution in the water subphase and so it doesn’t influence the
surface pressure measurements. It should be also volatile enough to quickly evaporate from the
surface after spreading, leaving behind the adsorbed block copolymer molecules. Chloroform
was the solvent of choice in this dissertation except in Chapter 2 where the use of a more polar
33
chloroform/ethanol mixture was necessary to dissolve the polystyrene-b-poly(acrylic acid)
dendrimer-like block copolymer sample. Other commonly used solvents include n-hexane,
benzene, diethyl ether, as well as mixtures such as for instance hexane-ethanol, benzene-ethanol,
or chloroform-methanol.87 After spreading, the surface active block copolymers expand to form
an adsorbed monolayer that quickly covers the entire available area present between the barriers.
In addition to the A/W interface which is the most commonly studied interface, others are
possible such as oil/water interfaces or gas (different from air)/water interfaces.
Figure 2-2. A Wilhelmy plate partially immersed in a water subphase.
The surface pressure measurements were achieved by use of the Wilhelmy plate technique
as shown in Figure 2-2.87 The surface tension (π) is defined by Equation 2-1,
γγγπ ∆−=−= 0 (2-1)
where γ0 represents the surface tension of the water subphase in the absence of an adsorbed
monolayer, and γ represents the surface tension of the water subphase in the presence of an
adsorbed monolayer. Water is a very interesting subphase when investigating monolayers since it
has an exceptionally high surface tension (≈ 73 mN/m at 20 oC and under atmospheric
pressure87) that allows a wide range of surface pressures to be obtained. When the Wilhelmy
plate is partially immersed into the subphase, the forces acting consist of gravity and surface
34
tension downward as well as an upward buoyancy due to displaced water. The resulting force F
is monitored and related to the surface tension γ as shown in Equation 2-2:
ghwtwtglwtF sp ρθγρ −++= cos)(2 (2-2)
where ρp and ρs correspond to the densities of the Wilhelmy plate and the subphase, respectively,
g corresponds to the gravitational constant, t (thickness), w (width), and l (length) are the
dimensions of the Wilhelmy plate, θ is the contact angle made by the subphase onto the plate (θ
= 0 for a completely wetted plate, cosθ = 1), and h corresponds to the plate height immersed into
the subphase. When measuring the change in F (∆F) for a stationary plate after formation of an
adsorbed Langmuir monolayer, assuming that t is negligible compared to w, the equation can be
reduced to:
γ∆=∆ wF 2 (2-3)
and therefore, the surface pressure π is directly related to the change in force and the width of the
plate:
wF
2∆
−=π (2-4)
With the Wilhelmy plate technique, the sensitivity can be as low as 5x10-3 mN/m. Nevertheless,
since very small amounts of impurities can affect surface pressure measurements, the water
subphase consists of Millipore filtered water with resistivities greater than 18.2 MΩ.cm.
Moreover, the trough needs to be carefully cleaned for instance with ethanol before forming a
Langmuir monolayer and rinsed several times with Millipore filtered water. Contamination with
impurities coming from the air can also be minimized by covering the trough or by carrying out
the experiments in a clean room. The subphase temperature can also be regulated with a
circulating water bath.
35
Once a Langmuir monolayer is formed at the A/W interface, a variety of surface pressure
related experiments are possible. The simplest experiment is the isotherm, where the monolayer
is compressed toward the center of the trough and the surface pressure recorded versus surface
density or mean molecular area (MMA, average area occupied by one molecule at the interface,
calculated from Equation 2-5).
A
n
CVNAM
MMA = (2-5)
A is the area between the barriers, Mn is the molecular weight of the sample, NA is the
Avogadro’s number, and C and V are the concentration and the volume of the spreading solution,
respectively. Upon compression, the adsorbed molecules move closer to one another, the surface
pressure increases, and the phase transitions taking place in the monolayer are characterized by
pseudoplateaus or inflection points in the surface pressure versus MMA (or surface density)
isothermal plot as exemplified in Figure 2-3 for small amphiphilic molecules. The monolayer
undergoes phase transitions from the gaseous state (G) to the liquid state (L) and eventually to
the solid state (S) upon high compression.
Figure 2-3. Schematic isotherm for small amphiphilic molecules.14
36
Other possible experiments include compression-expansion hysteresis, isobaric, or
isochoric experiments. In the hysteresis experiment, the barriers compress the monolayer up to a
target surface pressure (or target MMA) and expand several times. In the isobaric experiment,
the monolayer is compressed up to a desired surface pressure which is then kept constant by
small barriers adjustments, and the evolution of the MMA versus time is recorded. In the
isochoric experiment, the barriers are stopped after reaching the desired surface pressure (or
MMA), and the surface pressure evolution is subsequently recorded versus time. These last 3
experiments are usually used to investigate monolayer stability versus time or its formation
reversibility as further exemplified in Chapters 3, 4, and 5.
2.2 Langmuir-Blodgett Films and Atomic Force Microscopy
One of the interesting properties of the Langmuir trough is that it is easy to transfer a
Langmuir monolayer prepared at the A/W interface onto a solid substrate for further analysis.
The most common technique for monolayer transfer is done by vertically lifting a substrate into a
Langmuir monolayer compressed at the desired surface pressure which is kept constant by small
barrier adjustments (Figure 2-4). The resulting transferred film is called a Langmuir-Blodgett
(LB) monolayer. The substrate used in this dissertation is freshly cleaved hydrophilic mica
because it mimics the water subphase during transfer. It is also possible to prepare LB films
consisting of multilayers by successively transferring several monolayers. This technique
provides thin films of supramolecular assemblies with well-defined molecular arrangement,88
chemical composition, and thickness, and with fewer defects than if prepared directly from the
bulk.
The transfer ratio measures the degree of monolayer transfer. During LB monolayer
formation, part of the molecules initially adsorbed at the A/W interface are slowly transferred
onto the substrate, which results in a lower surface density in the Langmuir monolayer. To keep
37
the surface pressure constant during transfer (and therefore the surface density), the barriers have
to move closer together which therefore reduces the available area between them. The transfer
ratio can be subsequently directly calculated by dividing this decrease in surface area between
the barriers by the total surface area of the substrate. Transfer ratio values smaller or larger than
unity indicate that the packing of the transferred molecules is less or more dense than at the A/W
interface, respectively.
Substrate (mica)
Wilhelmy plate Barrier
Water
Barrier
Figure 2-4. LB film transfer onto a hydrophilic mica substrate.
The experimental technique used in this dissertation to examine the morphology of the LB
films in Chapters 3, 4, and 5 is atomic force microscopy (AFM). This technique provides a
topographical map of the LB films by scanning their surfaces with a sharp tip attached at the end
of a flexible cantilever which is deflected as it moves up and down mountains and valleys. A
laser is reflected off the end of the cantilever, and the deflection (measured by a photodiode) is
converted into a topographical image with a resolution of 0.1 nm accuracy. Contact and non-
contact AFM modes were first designed, but while in the contact mode soft samples can be
distorted, in the non-contact mode the tip can easily become stuck in the water layer covering the
38
surface of samples exposed to the atmosphere. The AFM tapping mode used in this dissertation
overcomes these problems. The tip is vibrated with sufficient energy to overcome the surface
tension of the contaminant water layer, and this vibration also prevents frictional forces from
dragging the sample, which is crucial when working with soft polymeric samples. The tapping
mode electronic setup is shown in Figure 2-5. As scanning occurs in three dimensions, the
scanner tube contains three piezo electrodes (a piezoelectric material is a substance that
proportionally contracts and expands depending on an applied voltage) for the X, Y, and Z
directions. In addition to its convenience in characterizing LB monolayers, AFM was also used
to image the molecularly imprinted nanoparticles after their deposition onto mica substrates as
further described in Chapter 7.
Figure 2-5. The AFM tapping mode electronic setup.89
39
2.3 Transmission Electron Microscopy and Quasi-Elastic Light Scattering
Transmission electron microscopy (TEM) and quasi-elastic light scattering (QELS) were
used in Chapters 6 and 7 to determine the shape and size of the core-shell nanocapsules and of
the molecularly imprinted nanoparticles.
Electron microscopes use a beam of highly energetic electrons to examine objects on a
very fine scale and can yield information such as sample topography, morphology, composition,
or crystallinity.90 The electron source first forms a beam of electrons that are accelerated toward
the specimen with a positive electrical potential. The beam is focused onto the sample, and
various interactions can occur inside the irradiated sample, affecting the electron beam as shown
in Figure 2-6.
Figure 2-6. Specimen interactions in electron microscopy.90
The electrons of interest in TEM are only the unscattered electrons, even though the other
types of electrons can be of interest for other applications. These unscattered electrons are
40
transmitted through the thin sample without any interaction occurring inside the sample. The
thicker areas will have fewer transmitted unscattered electrons and will therefore appear darker
than the thinner areas. The unscattered electrons strike a screen where light is generated,
allowing the user to see the sample image. Staining agents such as osmium tetroxide, ruthenium
tetroxide, or uranyl acetate can also be used to increase the contrast of thin samples that do not
sufficiently scatter the electron beam.
Quasi elastic light scattering (QELS, also called dynamic light scattering, DLS) is used to
determine the hydrodynamic radius or diameter of nano-sized particles suspended in a solution.
These particles are subject to Brownian motion, with small particles diffusing faster than large
particles according to the Stokes-Einstein equation (2-6) relating diffusion coefficient (D) and
particle radius (R):
RTk
D B
πη6= (2-6)
where kB is the Boltzmann constant, T is the temperature, and η is the solution viscosity. A beam
of laser light is first focused on the sample, the particles scatter the incoming light in all
directions, and the scattered photons are measured by a photomultiplier tube. The time variation
of the scattered intensity is then analyzed with a digital correlator used to compute the following
autocorrelation function C(τ):
BAeC += Γ− ττ 2)( (2-7)
where A and B are instrumental constants. The diffusion coefficient D is then determined from
the following 2 equations:
λ
θπ )2
sin(4 nq = (2-8)
41
2qD Γ
= (2-9)
where q is the scattering vector, n is the refracting index, θ is the scattering angle, λ is the laser
wavelength, and the particle hydrodynamic radius R is calculated from Equation 2-6. Each
monodisperse population of particle sizes produces its own unique autocorrelation function
(exponential decay). Mixtures of more than one size population produce sums of exponentials,
and therefore available algorithms can be used to extract “true” size distributions from complex
samples.91
2.4 Cyclic Voltammetry
Cyclic voltammetry was used in Chapter 6 to monitor the uptake versus time of
hydrophobic electrochemical probes inside the oil-core of the core-shell nanocapsules. In cyclic
voltammetry, the potential of a working electrode is continuously changed as a linear function of
time, with the rate of change of potential referred to as the scan rate. The potential is first
scanned in one direction and then reversed at the end of the first scan. The advantage of such a
technique in electrochemistry is that the product of the electron transfer that occurred in the
forward scan can be probed again in the reverse scan. Figure 2-7 shows the basic shape of the
current response for a cyclic voltammetry experiment. The bulk solution first contains only the
reduced form of the redox couple, so for potentials (in A) lower than the redox potential, there is
no net conversion of the reduced form into the oxidized form. As the potential is increased
toward the redox potential, a net anodic current appears and increases exponentially. As the
reduced form is oxidized, concentration gradients appear, and diffusion occurs down these
gradients. At point B (anodic peak) and beyond, the potential is sufficiently high so the reduced
species reaching the electrode are immediately oxidized, and the current therefore now depends
upon the rate of mass transfer which results in an asymmetric peak shape. In C, the potential is
42
reversed (decreased), and the current continues to decay. As the redox potential is approached, a
net reduction of the oxidized species takes place resulting in the appearance of a peak-shaped
cathodic current (D).
Figure 2-7. Basic shape of the current response for a cyclic voltammetry experiment.92
In the presence of a reversible redox system, that is for fast electron transfer processes at
the electrode (reaction kinetics fast compared to the scan rate), the surface concentrations of the
oxidized and the reduced species are maintained at the values required by the Nernst equation (2-
10):
)ln(r
oo
CC
nFRTEE += (2-10)
where E is the electrode potential, Eo is the standard potential of the redox couple, R is the
universal gas constant (8.314510 J K-1 mol-1), T is the temperature in K, n is the number of
electrons transferred in the half reaction, F is the Faraday constant (9.6485309x104 C mol-1), and
43
Co and Cr represent the surface concentrations of the oxidized and reduced species, respectively.
The peak potential separation (Epa-Epc) is equal to 57/n, the peak width is equal to 28.5/n, the
peak current ratio (Ipa/Ipc) is equal to unity, and finally the faradaic current (Ip) is given by the
following Randles-Sevcik equation (2-11):
CADnIp 2/12/12/351069.2 ν×= (2-11)
where A is the electrode area (cm2), D is the diffusion coefficient of the electroactive species
(cm2/s), ν is the scan rate (V/s), and C is the bulk concentration of the electroactive species
(mol/cm3). If A and ν are kept constant during successive cyclic voltammetry scans, a decrease in
the bulk concentration of the electroactive species can be simply measured by monitoring the
decrease in the Faradaic current Ip. This strategy was applied as further discussed in Chapter 6 to
measure the encapsulation abilities and kinetics of two electrochemical probes with reversible
electrochemical activities (ferrocene methanol and ferrocene dimethanol) inside the oil core-
silica shell nanocapsules.
44
CHAPTER 3 POLYSTYRENE-b-POLY(TERT-BUTYLACRYLATE) AND POLYSTYRENE-b-
POLY(ACRYLIC ACID) DENDRIMER-LIKE COPOLYMERS: TWO-DIMENSIONAL SELF-ASSEMBLY AT THE AIR-WATER INTERFACE
3.1 Introduction
As introduced in Chapter 1, block copolymers are of great interest for nanotechnological
applications because of their potential in forming well-defined morphologies in bulk,93 in
solution,94 or at interfaces.95 Amphiphilic block copolymers offer good opportunities to control
interfacial properties, and since they self-assemble at hydrophilic/hydrophobic environments
such as the air/water (A/W) interface to form regular arrangements of two-dimensional surface
micelles, they are particularly interesting as precursors for the synthesis of well-defined patterns
with feature sizes typically in the nanometer scale order. The nanosized structures can be easily
controlled by varying various parameters such as the nature of the blocks, the molecular weight,
the relative lengths of the blocks, or even the block copolymer architecture. With a view toward
understanding the relationships between polymer architecture and A/W interfacial aggregation, a
variety of polystyrene-b-poly(ethylene oxide) block copolymers with simple and complex
architectures (linear, star-shaped, mikto-arm, dendrimer-like…) has been synthesized and
extensively investigated in the past few years.22,96
Concerning polystyrene-b-poly(tert-butylacrylate) (PS-b-PtBA) and polystyrene-b-
poly(acrylic acid) (PS-b-PAA) block copolymers, only little work has been reported on their
A/W interfacial behavior. Lennox and co-workers were to our knowledge the only ones to report
on the two-dimensional surface micelle formation for PS-b-PtBA block copolymers,
investigating linear samples with relatively small PtBA chains compared to the PS block.24,97 The
self-assembly of PS-b-PAA block copolymers was also studied almost exclusively for linear
samples,20,98 and it is only very recently that Tsukruk and co-workers described the self-assembly
45
of a twelve-arm heteroarm star copolymer with alternating PS and PAA arms at the A/W
interface.99 In a recent publication, we described the synthesis of a variety of dendrimer-like
copolymers based on polystyrene and poly(tert-butylacrylate) or poly(acrylic acid), and a
preliminary investigation of their surface properties confirmed that such systems are ideal
candidates for the synthesis of Langmuir and Langmuir-Blodgett (LB) monolayers with well-
defined morphologies.100
In this chapter, we study the A/W interfacial behavior of two dendrimer-like copolymers
composed of an eight-arm star polystyrene core with a sixteen-arm poly(tert-butylacrylate) or
poly(acrylic acid) corona (Figure 3-1) by surface pressure measurements (isotherms, isochores,
and compression-expansion hysteresis experiments) and atomic force microscopy (AFM)
imaging. Among the different dendrimer-like copolymer samples previously synthesized, we
deliberately chose to focus here on these 2 samples since they have the largest number of arms
and the largest PtBA (or PAA) weight %. These polymers are significantly different from the
block copolymers studied in the previous literature in terms of architecture and chain length
ratio, which should therefore lead to an interesting and peculiar interfacial behavior.
Figure 3-1. Schematic sketch of the PS-b-PtBA (Dend1) and the PS-b-PAA (Dend2) samples.
46
3.2 Results and Discussion
3.2.1 PS-b-PtBA Dendrimer-Like Copolymer (Dend1)
The isotherm of Dend1 is presented in Figure 3-2, and the inset shows the static elastic
modulus plot (εs, Equation 3-1) versus surface pressure.101
)()(
MMAddMMAs
πε −= (3-1)
The surface pressure in the isotherm increases in the high MMA region with a behavior
characteristic of liquid expanded and liquid condensed phases until a plateau suggesting a
biphasic state is observed at 24 mN/m. Upon further compression in the low MMA region, the
pressure sharply increases up to values as high as 70 mN/m. PS and PtBA are two hydrophobic
polymers, but contrary to PS, the presence of slightly polar ester groups in PtBA allows the
adsorption of PtBA homopolymers and PtBA-based block copolymers at the A/W interface. The
isotherm of Dend1 is very similar in shape to the isotherms of PtBA homopolymers, with only
one phase transition observed at 24 mN/m.102 Moreover, the extrapolation of the linear portion of
the isotherm below the plateau to zero pressure (Apancake)103 yields a value of 29 Å2 per PtBA
repeat unit, which is in good agreement with the previously reported literature.104 These two
observations indicate that Dend1 behaves similarly in terms of surface activity to PtBA
homopolymers, with the PS blocks desorbed in the air on top of the adsorbed PtBA blocks.
Nevertheless, these results are significantly different from those obtained by Lennox and co-
workers for linear PS-b-PtBA diblock copolymers with relatively low PtBA wt %.24,97 They
showed that the presence of the PS block induced two additional phase transitions of the PtBA
block below 20 mN/m and suggested that these transitions might originate from the peculiar
surface aggregation into circular micelles. As shown in the inset of Figure 3-2, only one local
minimum characteristic of a phase transition is observed at 24 mN/m for the static elastic
47
modulus versus surface pressure plot. Moreover, Dend1 also aggregates into spherical surface
micelles (Figure 3-3). This indicates that in the case of circular surface micelles, the extra phase
transitions are observable only for small PtBA chain length over PS chain length ratios (high PS
wt %). In the case of Dend1, we believe these transitions might be present but cannot be detected
since they involve only the small amount of PtBA repeat units present in the vicinity of the PS
micellar core, with the majority of the PtBA repeat units being too far and behaving similarly as
in PtBA homopolymers.
Figure 3-2. Isotherm of Dend1. (Inset) Static elastic modulus plot versus surface pressure.
The AFM images of the LB films transferred below the plateau at 5, 10, 15, and 20 mN/m
(Figures 3-3a, 3-3b, 3-3c, and 3-3d) show the circular micellar aggregation of Dend1 mentioned
before with the bright circular domains corresponding to the desorbed PS cores (~ 1-1.5 nm
thick) and the darker background corresponding to the adsorbed PtBA chains. Estimation of the
number of Dend1 molecules per circular micelle by the total area method105 yielded an average
48
aggregation number of 3, independent of the surface pressure. PS has a high glass transition
temperature (Tg ~ 100 oC for high molecular weight PS),106 and therefore, once a surface micelle
has formed during the solvent spreading process, the micellar cores come closer to each other
during compression but the aggregation number remains unchanged with the chains stuck within
a particular micelle (the temperature of the experiment is well below the Tg of the desorbed PS
cores). Compression-expansion hysteresis experiments were carried out with target surface
pressures below the plateau and all the compression and expansion curves overlapped, which is
indicative of film formation reproducibility and thermodynamic stability. AFM imaging of the
LB films prepared during the second compression step of the hysteresis experiment did not
reveal any significant difference compared to the LB films formed during the first compression.
The aggregation number is much smaller than the value of 95 block copolymers per circular
surface micelle reported by Lennox and co-workers for a linear PS305-b-PtBA222 diblock
copolymer,97 which was expected based on the previous studies investigating the relationship
between aggregation number and block copolymer architecture. For instance, increasing the wt
% of the adsorbed PEO segment in linear PS-b-PEO diblock copolymers resulted in a lower
aggregation number.96c Even lower aggregation numbers were observed for PS-b-PEO three-arm
stars with a PEO corona compared to the linear analogues or compared to the reverse star
architecture with a PS corona.22 Similarly to these examples, the adsorbed PtBA chains of Dend1
are present in the corona and are approximately 10 times longer than the desorbed PS chains.
Moreover, the large number of arms connected to the central calix[8]arene core and the
dendrimer-like structure with 2 PtBA chains attached at the end of each PS arm also heavily
favor the formation of highly curved interfaces,96f therefore also considerably reducing the
aggregation number.
49
Figure 3-3. Topographic AFM images of Dend1 LB films transferred at 5 (a), 10 (b), 15 (c), 20
(d), 24 (middle of plateau, MMA = 20,000 Å2, e and f), and 40 mN/m (g and h).
A LB film of Dend1 was also transferred in the middle of the plateau (MMA = 20,000 Å2),
and the resulting AFM images are shown in Figures 3-3e and 3-3f. In this case, no transfer ratio
could be calculated since the barriers were manually stopped after compression before transfer,
and therefore these two AFM images only qualitatively reflect the A/W interfacial aggregation.
Nevertheless, the images clearly confirm the biphasic state of the plateau, with areas exhibiting
circular surface micelles similarly as for lower pressures and other areas with aggregated
domains significantly thicker (~ 3 nm) than the PS micellar cores. It was previously suggested
that at the beginning of the plateau region, the tert-butyl groups reorient from prone to vertical,
with further compression leading to film collapse.107 Since PtBA is a hydrophobic polymer, it
collapses by aggregating on top of the water surface. The aggregated areas observed in Figures
3-3e and 3-3f therefore likely correspond to desorbed Dend1 mono or multilayers. Compression-
expansion hysteresis experiments (3 cycles) were carried out in the plateau region, and the
resulting curves are shown in Figure 3-4. All the compression curves closely overlapped,
indicating that the collapsed films are able to go back to their original adsorbed monolayer state
50
upon barrier expansion. AFM imaging of the LB films transferred during the second
compression at a surface pressure of 20 mN/m confirmed the presence of adsorbed and circular
surface micelles and the absence of collapsed domains. Nevertheless, a drop in pressure is
observed during barrier expansion. Faster barrier speeds increased the pressure drop, which
suggests that the readsorption kinetics of the collapsed PtBA segments are slower than the barrier
expansion speed.
Figure 3-4. Compression-expansion hysteresis plot of Dend1 (target MMA = 20,000 Å2).
Further compression in the low MMA region leads to a sharp surface pressure increase.
The films formed in this region of the isotherm should be unstable since the surface pressure is
higher than the PtBA collapse pressure of 24 mN/m. This was checked by carrying out isochoric
experiments. A typical isochore after compression up to 40 mN/m is shown in Figure 3-5. The
surface pressure undergoes a sudden drop within the first seconds of the experiment before
leveling off after a few minutes around 24 mN/m. It is not clear to us whether the sharp surface
pressure increase upon high compression in the isotherm originates from interactions between
the collapsed and desorbed aggregates formed in the plateau region or between eventual
remaining adsorbed PtBA segments. In any case, the isochoric experiments confirmed that the
51
films formed above the plateau are not thermodynamically stable and relax within minutes down
to the plateau surface pressure of 24 mN/m, which can be also seen as the equilibrium spreading
pressure of the PtBA blocks. Compression-expansion hysteresis experiments (3 cycles) were also
carried out in this high surface pressure region, and an example for a target pressure of 40 mN/m
is shown in Figure 3-6. The results are essentially similar to the hysteresis experiments carried
out in the middle of the plateau, with the compression curves overlapping each other, and the
expansion curves exhibiting a drop in pressure that corresponds to the slow readsorption of the
collapsed PtBA chains. AFM imaging of the LB films transferred during the second compression
at a surface pressure of 20 mN/m also confirmed here the presence of circular surface micelles
only. The morphology of the LB film transferred at 40 mN/m is shown in Figures 3-3g and 3-3h,
and aggregated domains in addition to the circular surface micelles are observed similarly as
within the plateau. Nevertheless, the transfer ratio was in this case significantly smaller than one
(~ 0.25), which indicates that the film underwent non-negligible expansion during transfer,
suggesting that the density of desorbed aggregates on the water surface is significantly greater
than as shown in Figures 3-3g and 3-3h.
Figure 3-5. Surface pressure/time isochoric relaxation plot of Dend1 after compression up to 40 mN/m.
52
Figure 3-6. Compression-expansion hysteresis plot of Dend1 (target pressure = 40 mN/m).
3.2.2 PS-b-PAA Dendrimer-Like Copolymer (Dend2)
As a preliminary study, we investigated the behavior of a PAA homopolymer (PAA250K),
since to our knowledge no information is available in the previous literature on the Langmuir and
LB monolayers of PAA homopolymers. PAA250K could not be dissolved in chloroform, so a
chloroform/ethanol mixture (1/2 in volume) was used instead as the spreading solvent. PAA is a
weak acid (pKa ~ 5.5),108 and its degree of ionization and consequently its surface activity are
therefore strongly pH dependent. Under basic conditions (pH = 11), the acid groups of PAA are
deprotonated, and stable Langmuir monolayers of PAA250K could not be prepared, with the
surface pressure remaining close to 0 mN/m during barrier compression. When ionized, PAA is
not surface active109 and irreversibly dissolves into the aqueous subphase. Under acidic
conditions (pH = 2.5), the acid groups of PAA are protonated, the chains become less
hydrophilic and surface active, and therefore stable Langmuir monolayers of PAA250K could now
be prepared. The resulting isotherm is shown in Figure 3-7. The surface pressure slowly
increases in the high MMA region before a plateau is reached with the surface pressure leveling
53
off around 3.5 mN/m. In the high MMA region, the surface pressure increases with a behavior
characteristic of liquid expanded and liquid condensed phases, and PAA250K is likely adsorbed in
a pancake conformation. The AFM image shown in Figure 3-8a (π = 1.5 mN/m) confirmed the
presence of a pretty homogeneous PAA250K LB film with the randomly distributed darker holes
possibly forming after transfer as the PAA chains un-swell (dehydrate) during the drying step.
PAA in its protonated form is surface active but it is still highly water soluble, and it has been
previously shown for aqueous solutions of protonated PAA that the surface pressure reaches a
maximum, for instance around 3 mN/m for a molecular weight similar to PAA250K.109a This value
correlates well with the plateau surface pressure, which indicates that, at this pressure, PAA250K
desorbs from the interface and aggregates inside the water subphase. The AFM images obtained
for the LB films transferred at 3 and 3.5 mN/m are shown in Figures 3-8b and 3-8c, respectively.
As the films are compressed within the plateau region, more and more desorbed aggregates (~ 3-
5 nm thick) are observed. This indicates that when the PAA chains collapse by dissolving into
the water subphase, they stay in the vicinity of the interface, probably remaining anchored by
some adsorbed acrylic acid repeat units that have not collapsed yet.
Figure 3-7. Isotherm of PAA250K (pH = 2.5).
54
Figure 3-8. Topographic AFM images of PAA250K LB films transferred at 1.5 (a), 3 (b), and 3.5 mN/m (c).
The spreading solutions of Dend2 were prepared by using the same solvent mixture as for
PAA250K (chloroform/ethanol mixture, 1/2 in volume). Similarly as for PAA250K, no stable
Langmuir monolayers could be formed for Dend2 under basic conditions with its irreversible
dissolution in the aqueous subphase. Contrary to the other PS-b-PAA systems reported in the
previous literature,20,98,99 the PS wt % in Dend2 is too low to allow it to act as an anchor and thus
preventing the dendrimer-like aqueous dissolution. Nevertheless, under acidic conditions (pH =
2.5), the PAA segments of Dend2 are now protonated, and, similarly as for PAA250K, Dend2
becomes surface active. The resulting isotherm is shown in Figure 3-9, and the inset shows the
monolayer compressibility plot (K, Equation 3-2) versus surface pressure.96g
)()(11
πε dMMAd
MMAK
s
−== (3-2)
In the high MMA region, the surface pressure slowly increases before reaching a pseudoplateau
around 5 mN/m (maximum in monolayer compressibility as seen in the inset of Figure 3-9) and
sharply increasing under high compression up to approximately 10 mN/m. A pseudoplateau was
already observed in the previous literature by Currie et al. only for long PAA chains in linear PS-
b-PAA systems under acidic conditions, where they demonstrated by surface pressure and
ellipsometry measurements the presence of a pancake-to-brush transition in the pseudoplateau
55
region with the PAA segments dissolving into the water subphase.20 It is not very clear at this
point why the pseudoplateau surface pressure is a little bit higher for Dend2 (5 mN/m) compared
to the PS-b-PAA copolymers studied by Currie et al. or compared to PAA250K (3-4 mN/m). It
might simply be a PAA chain length dependence, or possibly a peculiar influence of the
dendrimer-like architecture.
Figure 3-9. Isotherm of Dend2 (pH = 2.5). (Inset) Compressibility plot versus surface pressure.
We investigated here the morphological evolution by AFM imaging as the Langmuir films
of Dend2 are compressed, and the resulting images of the LB films are shown in Figure 3-10. At
low surface pressure (π = 2 mN/m, Figure 3-10a), Dend2 aggregates into spherical surface
micelles with bright PS cores (~ 1 nm thick) and a darker background corresponding to the PAA
blocks adsorbed in a pancake conformation similarly as in Figure 3-8a for PAA250K. The average
aggregation number was estimated around 5 by the total area method.105 Such a low value can be
56
rationalized similarly as for Dend1, with the long PAA chains present in the corona, the
dendrimer-like shape, and the large number of arms emanating from the central calix[8]arene
core all heavily favoring the formation of highly curved interfaces. The images shown in figures
3-10b through 3-10j were recorded for surface pressures within the pseudoplateau and ranging
from 4 to 8 mN/m. As the films are compressed, the PAA segments progressively dissolve in the
aqueous subphase, underneath the PS cores of the surface micelles that aggregate into larger and
thicker (~ 2 nm) domains. At the end of the pseudoplateau, in the region where the surface
pressure sharply increases under high monolayer compression (π = 8 mN/m, Figures 3-10k and
3-10l), all the PAA chains have desorbed in the water subphase and stretch to form a brush
underneath the aggregated PS cores. Contrary to the hydrophobic PtBA segments of Dend1 that
collapsed above the A/W interface, the hydrophilic PAA segments of Dend2 anchored by the PS
segments dissolve into the water subphase. Compression-expansion hysteresis experiments were
also conducted within the pseudoplateau region (target pressure = 5 mN/m), and the results are
presented in Figure 3-11. Very little hysteresis is observed with the pseudoplateau still present
after numerous cycles, which means that the desorbed PAA segments can return to their original
adsorbed state at low pressure after monolayer decompression. This was verified by transferring
a LB film at 2 mN/m during the second compression cycle, and AFM imaging revealed the
presence of circular surface micelles with no aggregated domains. We do not have yet at this
point enough experimental evidence to rationalize the small hysteresis shift, but it could for
instance come from some entanglement of the PAA chains during the pancake-to-brush
transition or from a different arrangement of the expanded PAA chains in terms of hydration and
conformation compared to the one adopted after spreading and before the first compression.110
57
Figure 3-10. Topographic AFM images of Dend2 LB films transferred at 2 (a), 4 (b), 4.5 (c and d), 5 (e and f), 5.5 (g and h), 6 (i and j), and 8 mN/m (k and l).
Figure 3-11. Compression-expansion hysteresis plot of Dend2 (target pressure = 5 mN/m, pH = 2.5).
58
3.3 Conclusions
In this work, the A/W interfacial behavior of two dendrimer-like block copolymers based
on polystyrene and poly(tert-butylacrylate) (Dend1) or poly(acrylic acid) (Dend2) and their LB
film morphologies on hydrophilic mica substrates were investigated. Dend1 formed
thermodynamically stable Langmuir monolayers and self-assembled into circular surface
micelles up to 24 mN/m with a very low aggregation number (~ 3), likely resulting from the high
PtBA wt % and from the peculiar dendrimer-like architecture. At 24 mN/m, the PtBA segments
desorbed and aggregated on top of the water surface, and all the monolayers formed beyond this
threshold were metastable and relaxed down to 24 mN/m. Poly(acrylic acid) is surface active
only under acidic pH conditions below its pKa value (~ 5.5), and a preliminary study on a PAA
homopolymer (PAA250K, Mn = 250,000 g/mol) showed that it is able to form stable monolayers
up to ~ 3.5 mN/m before dissolving in the water subphase. Because of its small PS wt %, Dend2
similarly did not form stable Langmuir monolayers under high pH conditions, and therefore its
self-assembly at the A/W interface was investigated only under acidic conditions. The isotherm
indicated the presence of a pseudoplateau at 5 mN/m characteristic of a phase transition that
corresponds to a pancake-to-brush transition, with the progressive aqueous dissolution of the
PAA segments underneath the anchoring PS cores. For pressures below the pseudoplateau,
Dend2 molecules also aggregated into circular surface micelles with a very low aggregation
number (~ 5). This study confirmed that various parameters such as for instance polymer
architecture, chain length/polarity, surface density, or even subphase pH can all strongly
influence the self-assembly of block copolymers at the A/W interface. Therefore, while this
work gave some hint of the interesting interfacial properties of these two novel PS-b-PtBA and
PS-b-PAA dendrimer-like copolymers, additional investigations with other samples with
59
different numbers of arms and chain lengths are necessary for a better understanding of the
influence of the dendrimer-like structure.
3.4 Experimental Methods
3.4.1 Materials
The synthesis of the PS-b-PtBA (Dend1, Mn = 240,000 g/mol, PDI = 1.26) and PS-b-PAA
(Dend2, Mn = 139,000 g/mol, PDI = 1.28) dendrimer-like copolymers was realized according to
a previously reported procedure.100 These block copolymers consist of an eight-arm PS core (Mn
= 10,000 g/mol, ~ 12 styrene units/arm) with a sixteen-arm PtBA (Mn = 230,000 g/mol, ~ 112
tert-butylacrylate units/arm, PS wt % ~ 4 %) or PAA (Mn = 129,000 g/mol, ~ 112 acrylic acid
units/arm, PS wt % ~ 7%) corona. The PAA homopolymer (PAA250K, Mn ~ 250,000 g/mol) was
purchased from Aldrich Chemical Co. and was used as received without further purification.
3.4.2 Langmuir Films
Surface pressure measurements were accomplished by use of a Teflon Langmuir trough
system (W = 160 mm, L = 650 mm, KSV Ltd., Finland) equipped with two moving barriers and a
Wilhelmy plate. The PS-b-PtBA dendrimer-like copolymer sample (Dend1) was prepared by
dissolving approximately 1 mg of polymer in 1 mL of chloroform. The PS-b-PAA dendrimer-
like copolymer (Dend2) and the PAA homopolymer (PAA250K) samples were prepared by
dissolving approximately 1 mg of polymer in 1 mL of a chloroform/ethanol mixture (1/2 in
volume). Volumes ranging from 10 to 30 µL were spread dropwise with a gastight Hamilton
syringe on Millipore filtered water (subphase resistivity ≥ 18.2 MΩ.cm) of the desired pH value,
and the spreading solvents were allowed to evaporate for 30 min. In all the experiments,
subphase temperature and barrier speed were kept constant at 25 °C and 5 mm/min, respectively.
60
3.4.3 AFM Imaging
The LB films were formed by transferring the Langmuir films of Dend1, Dend2, and
PAA250K onto freshly cleaved mica at various surface pressures. The desired surface pressure
was attained at compression/expansion rates of +/-5 mm/min. Once the films had equilibrated at
a constant surface pressure for 15 min, the mica substrate was then pulled out of the water at a
rate of 1 mm/min. All the transfer ratios were close to unity unless otherwise stated, which is
indicative of successful transfer. The transferred films were air-dried in a desiccator for 24 h and
subsequently scanned in tapping mode with a Nanoscope III AFM (Digital Instruments, Inc.,
Santa Barbara, CA) using Nanosensors silicon probes (dimensions: T = 3.8-4.5 µm, W = 26-27
µm, L = 128 µm). All the images were processed with a second-order flattening routine.
61
CHAPTER 4 LANGMUIR AND LANGMUIR-BLODGETT FILMS OF POLY(ETHYLENE OXIDE)-b-
POLY(ε-CAPROLACTONE) STAR-SHAPED AND LINEAR BLOCK COPOLYMERS
4.1 Introduction
Poly(ε-caprolactone) (PCL) and poly(ethylene oxide) (PEO) are two biocompatible
polymers with PCL biodegradation leading to nontoxic products.111,112 Low molecular weight
PEO is not biodegradable but can be eliminated from the body by the renal system.113 The
introduction of a hydrophilic and highly flexible PEO block at the end of a PCL block allows the
potential to tailor PCL properties such as its high hydrophobicity, high crystallinity, and slow
biodegradation.114 Therefore, synthesis of PEO-b-PCL copolymers with various architectures has
resulted in a growing number of reports citing potential applications including drug delivery
systems, temporary bioabsorbable implants, and tissue engineering.112,115,116
The air/water (A/W) interface has received much attention to investigate the self-assembly
of various amphiphilic molecules because of its ability to mimic hydrophilic/hydrophobic
interfaces. When polymers having biological applications are considered, such as PEO and PCL,
the A/W interface is of particular interest because of its presence in many biological systems.
Both PEO and PCL homopolymers spontaneously adsorb at the A/W interface to form stable
monolayers in a moderate surface pressure range. PEO homopolymers have been extensively
studied in solutions and at interfaces, and it has been shown that PEO chains irreversibly dissolve
in the water subphase at relatively low surface pressures.117 Surprisingly, very little interest has
been put into fundamentally understanding the aggregation in Langmuir monolayers of PCL-
based block copolymers and PCL homopolymers. Leiva et al. were to our knowledge the first to
report on the A/W interfacial properties of PCL homopolymers, and they observed collapse
pressure values ranging from 13 to 20 mN/m depending on the molecular weight, and
aggregation of the hydrophobic PCL chains on top of the water surface after collapse.118 It is only
62
very recently that Esker and co-workers extensively studied PCL crystallization mechanism in
the collapse region by Brewster angle microscopy and isobaric area relaxation analyses.101,119
The only example in the previous literature investigating the A/W self-assembly of PCL-based
block copolymers was reported by Lee et al. for an architecture with two linear PCL chains
anchored to a dendritic hydrophilic head.120
Chemically attaching a hydrophilic PEO block to a hydrophobic PCL block enhances the
amphiphilicity and the surface activity of the resulting block copolymer, allowing higher surface
pressures to be reached, as previously shown for other PEO-based amphiphilic block
copolymers.22,23 To our knowledge, no work has been reported on the A/W interfacial behavior
of block copolymers based on PEO and PCL. With a view toward guest encapsulation, we
recently reported the synthesis of a series of poly(ethylene oxide)-block-poly(ε-caprolactone)
copolymers with star-shaped and linear architectures by ring-opening polymerization of ε-
caprolactone at the end of hydroxyl-terminated star-shaped or linear PEO macroinitiators.121 In
the present chapter, we study the behavior of these samples as well as the corresponding PEO
and PCL homopolymers at the A/W interface (Langmuir monolayers) by surface pressure
measurements (isotherms, compression-expansion hysteresis, and isobaric relaxation
experiments), and we employ AFM to characterize the Langmuir-Blodgett (LB) films’
morphologies after transfer onto hydrophilic mica substrates. The PEO-b-PCL five-arm stars
consist of a hydrophilic PEO core with 9 ethylene oxide units/arm with hydrophobic PCL chains
at the star periphery. Each star contains different amounts of PCL, varying from 0 to 18 ε-
caprolactone units/arm (Figure 4-1 and Table 4-1). The linear PEO-b-PCL diblock copolymers
synthesized from a linear PEO macroinitiator (PEO2670, Mn = 2,670 g/mol, ~ 60 ethylene oxide
repeat units) contain different amounts of PCL, varying from 11 to 35 ε-caprolactone repeat units
63
(Figure 4-2 and Table 4-2). The other PEO homopolymer PEO2000 (Mn = 2,000 g/mol; PDI =
1.20) and the PCL homopolymers PCL1250 (Mn = 1,250 g/mol; PDI = 1.50), PCL2000 (Mn =
2,000 g/mol; PDI = 1.45), and PCL10000 (Mn = 10,000 g/mol; PDI = 1.40) were commercially
available (purchased from Aldrich Chemical Co.) and were used as received without further
purification.
Figure 4-1. The star-shaped PEO-b-PCL block copolymers.
Table 4-1. Characteristic values of the star-shaped polymers.
Name PEO core Star#1 Star#2 Star#3 Star#4 Star#5 Star#6
Mn a (g/mol) 2,150 4,200 6,370 8,200 9,630 11,220 13,110
PDIb 1.10 1.25 1.37 1.35 1.45 1.36 1.44Avg no. of ethylene oxide
units per arm 9 9 9 9 9 9 9
Avg no. of ε-caprolactone units per arm 0 3 6 9 12 15 18
a Determined by 1H NMR b Determined by GPC calibrated with linear polystyrene standards.
64
Figure 4-2. The linear PEO-b-PCL block copolymers.
Table 4-2. Characteristic values of the linear PEO macroinitiator and of the linear diblock copolymers.
Name PEO2670 PEO60-b-PCL11
PEO60-b-PCL19
PEO60-b-PCL27
PEO60-b-PCL35
Mn a (g/mol) 2,670 3,940 4,780 5,780 6,680
PDIa 1.05 1.10 1.13 1.11 1.24Avg no. of ethylene
oxide units 60 60 60 60 60
Avg no. of ε-caprolactone units 0 11 19 27 35
a Determined by GPC calibrated with linear poly(ethylene oxide) standards.
4.2 Results and Discussion
4.2.1 PEO Homopolymers
The isotherms of the five-arm PEO core for subphase pH values of 5.5 (Millipore filtered
water) and 13 (0.1 M NaOH) are presented in Figure 4-3. The isotherms of the linear PEO
homopolymers (PEO2000 and PEO2670) are also included. The inset shows the same isotherms
normalized with respect to the number of ethylene oxide repeat units. For PEO2000 and
PEO2670, the surface pressure increases until a pseudoplateau that corresponds to the
irreversible aqueous dissolution of the PEO chains is reached at 5 and 6.2 mN/m, respectively.
As previously described, Langmuir films of PEO homopolymers are thermodynamically stable
for low surface pressures.117 Upon compression the monolayers collapse, and the water-soluble
PEO chains irreversibly dissolve in the water subphase for pressures that are molecular weight
dependent (maximum collapse pressure value ~ 10 mN/m for high molecular weight PEO).122
65
The behavior of the PEO core significantly differs from that of the linear analogue PEO2000. No
pseudoplateau is observed, and the pressure slightly increases only for high compression, as
indicated by the shift toward the low area per monomer region compared to PEO2000. The
Millipore water pH (5.5) is acidic enough to protonate the triamine present in the center of the
PEO core, therefore increasing its water solubility compared to the neutral form. To investigate
the pH influence on the surface activity of the PEO core, a control isotherm was carried out with
a subphase pH value of 13, where the triamine is no longer protonated. The isotherm shape was
very similar to that at pH = 5.5. A slightly increased pressure was obtained for high compression,
but no pseudoplateau was observed. The absence of a pseudoplateau and the small pressure
increase only for high compression indicate a negligible surface activity of the adsorbed
molecules and the presence of a nonstable film. This assumption was verified by carrying out
compression-expansion hysteresis experiments on the PEO core for a target pressure of 2 mN/m
(Figure 4-4). Successive compression/expansion curves shifted toward smaller mean molecular
areas, which is indicative of a loss of material in the aqueous subphase. The pressure increase is
thought to arise from a kinetic effect, where the barrier compression speed is faster than the
dissolution rate of the PEO core molecules into the aqueous subphase, leading to a metastable
monolayer. Compared to the linear analogue, the star-shaped PEO core not anchored to
hydrophobic chains has little or no surface activity at the A/W interface and is therefore
thermodynamically more stable being solvated in the aqueous subphase than being adsorbed at
the interface, independent of the surface pressure. This behavior can be related to the high
polarity (presence of five hydroxyl groups) and the compact architecture (star-shaped) of the
PEO core, which is therefore more likely to dissolve in the water subphase than the linear
analogue.
66
Figure 4-3. Isotherms of the PEO homopolymers. (Inset) Same isotherms normalized with
respect to the number of ethylene oxide units.
Figure 4-4. Compression-expansion hysteresis plot of the PEO core (target pressure = 2 mN/m).
67
LB films of the PEO core could not be prepared because of its Langmuir monolayer
instability, so only PEO2000 and PEO2670 were transferred at 2 mN/m. PEO crystallization
from the bulk and organic solvents has been extensively investigated under various experimental
conditions including in thin films.123,124 The AFM scan of the PEO LB films indicated the
presence of a flat surface with no PEO crystals. PEO is highly hydrophilic, and previous studies
showed that every ethylene oxide unit can be hydrated with up to three water molecules.125
Recent studies in PEO-b-PCL systems indicated that hydrated PEO chains essentially lost their
ability to crystallize.126 The fact that we did not observe any crystallization in PEO2000 and
PEO2670 LB films probably arose from a similar situation where as the films dried, the PEO
chains remained highly hydrated and could not crystallize. PEO also has a high affinity for the
hydrophilic mica substrate, which can be another energy barrier preventing aggregation and
chain folding necessary for crystallization to take place.
4.2.2 PCL Homopolymers
While the main objective of our study was to gain a deeper understanding of the A/W
interfacial behavior of PEO- and PCL-based block copolymers, a preliminary study on PCL
homopolymers involving isothermal and compression/expansion hysteresis experiments with
experimental parameters (barrier speeds and target pressures) similar to the ones used for the
linear and star-shaped PEO-b-PCL samples needed to be done. The isotherms of the PCL
homopolymers introduced earlier are presented in Figure 4-5, with a log scale on the x-axis for
convenient visualization. The inset shows the same isotherms normalized with respect to the
number of ε-caprolactone repeat units. At low pressures, the PCL chains are adsorbed at the
interface with a behavior characteristic of liquid expanded and liquid condensed phases. Below
the collapse pressure, theoretical investigations by Ivanova et al. showed that PCL chain packing
68
and orientation is probably surface pressure-dependent.127 The inset shows that below the
collapse pressure, chain packing is not molecular weight-dependent. Compression-expansion
hysteresis cycles were performed in this surface pressure range and all the compression-
expansion curves closely overlapped, confirming the thermodynamic stability of the
monolayers.128,129
Figure 4-5. Isotherms of the PCL homopolymers. (Inset) Same isotherms normalized with respect to the number of ε-caprolactone units.
The collapse region of most amphiphilic molecules is more difficult to quantitatively
analyze because it is a thermodynamically unstable pressure region where various phenomena
such as desorption of molecules from the interface or local formation of multilayers can happen.
Many polymers show viscoelastic behavior in this region, leading to surface properties that are
barrier speed-dependent.130 As shown in Table 4-3, the collapse pressure (defined as the point
where a sudden change in slope is observed) is molecular weight-dependent, with the longer
PCL chains collapsing at lower pressures. PCL is a highly hydrophobic polymer, and its collapse
69
therefore results in the aggregation of the water-insoluble PCL chains on top of the aqueous
subphase. An interesting feature characteristic of PCL collapse is the obvious decrease in
pressure after the collapse point, a phenomenon that was attributed by Esker and co-workers to
PCL crystallization directly on the water surface. Longer PCL chains have increased
hydrophobicity and crystallinity compared to smaller PCL chains, which explains the lower
collapse pressure and increased pressure drop after the collapse point.
Table 4-3. Collapse pressure values of the PCL homopolymers. Name PCL1250 PCL2000 PCL10000 Collapse pressure (mN/m) for
compression speed = 5 mm/min 14.6 12.9 11.1
Collapse pressure (mN/m) for compression speed = 100 mm/min N/A 14.1 11.4
Further evidence for PCL crystallization came from the isotherms recorded for a barrier
compression speed of 100 mm/min (Figure 4-6). The pressure decrease after the collapse point
almost completely vanishes, and as shown in Table 4-3 the collapse pressure increases. This
indicates that in this case the rate of crystallization is slow compared to the compression speed
and that the resulting monolayers are metastable in this pressure region. Thermodynamic
collapse pressure values could not be easily obtained experimentally with our equipment because
it would require infinitely slow compressions. The collapse pressure for PCL1250 (100 mm/min)
could not be accurately determined because its isotherm does not show a clear collapse, but the
shallower turning point between 12 and 15 mN/m can nevertheless be attributed to the collapse
“point” of this film.
70
Figure 4-6. Isotherms of the PCL homopolymers (compression speed = 100 mm/min).
Compression-expansion hysteresis cycles were also carried out beyond the collapse point.
Figure 4-7 shows the compression-expansion hysteresis experiment carried out for PCL1250.
The first compression is similar to the isotherm reported in Figure 4-5. As the film is expanded,
the pressure suddenly drops until an expansion pseudoplateau appears at a pressure significantly
lower than the collapse pressure during compression. This expansion pseudoplateau corresponds
to the readsorption/melting of the PCL chains that had previously crystallized. This
pseudoplateau pressure is molecular weight-dependent, with the smaller PCL chains readsorbing
at higher pressure (10, 8, and 4 mN/m for PCL1250, PCL2000, and PCL10000, respectively).
The compression curve of the second cycle is slightly shifted toward smaller areas, and a
decrease is observed for the collapse pressure. This phenomenon is the result of residual PCL
crystals that remained on the water surface from the initial monolayer compression and that act
as nucleation sites for the crystallization of the other adsorbed PCL chains.101,119 Further
71
compression-expansion cycles essentially overlapped the second cycle with no noticeable
pressure changes concerning the collapse or the expansion pseudoplateau.
Figure 4-7. Compression-expansion hysteresis plot of PCL1250.
Typical AFM images of the PCL homopolymers LB films transferred below monolayer
collapse (π = 7 mN/m), before crystallization on the water surface takes place, are shown in
Figure 4-8a and 4-8d. Surprisingly, PCL crystals could be observed. According to the isotherms
that are characteristic of liquid condensed phases in this low surface pressure range, the PCL
homopolymers are transferred into smooth and hydrated monolayers adsorbed onto the mica
surface. We believe that upon drying, during and after transfer, part of the PCL chains leave the
surface and crystallize, which likely results in a mica surface only partially covered with
adsorbed or crystallized PCL chains. For comparison, PCL homopolymers were also transferred
beyond monolayer collapse, after crystallization on the water surface takes place (Figure 4-8b, 4-
8c, and 4-8e). The transfer ratios were in this case significantly greater than 1 (~ 2-3), which was
predictable since crystallization and therefore intrinsic MMA decrease take place over time in
72
this surface pressure range. PCL crystals were also observable, with spherulitic architectures
significantly different from those obtained by Esker and co-workers directly at the A/W interface
using BAM.101,119 This suggests that the AFM images recorded above monolayer collapse are not
only the result of PCL crystallization at the A/W interface but also the result of additional PCL
crystallization taking place during and after transfer. By use of cross-section analysis (Figure 4-
8f), all the PCL crystals were determined to be approximately 7.5 nm thick, which is consistent
with the previously reported literature on PCL lamellae thickness.131 The thickness was
independent of PCL molecular weight, which indicates that the chains stretch perpendicular to
the surface and fold every 7.5 nm, with the crystals probably growing parallel to the surface.
Nevertheless, further comparison between PCL crystallization in LB films and the previously
reported work on PCL crystallization in bulk,132 from an organic solution,133 or even in thin
films134-136 remains difficult to make because different types of variables are involved.
Figure 4-8. Topographic AFM images of PCL homopolymers LB films transferred below and
above monolayer collapse: PCL2000 at 7 (a) and 13 mN/m (b and c), and PCL10000 at 7 (d) and 11.2 mN/m (e). (f) Cross-section analysis performed at the edge of a PCL2000 crystal.
73
4.2.3 Star-Shaped PEO-b-PCL Block Copolymers
The isotherms of the PEO-b-PCL five-arm stars are presented in Figure 4-9. They exhibit
essentially three different regions that can be attributed to different conformations of the polymer
chains. In the high MMA region, the surface pressure slowly increases as the films are
compressed until a pseudoplateau is reached for intermediate mean molecular areas. As the
compression continues in the low MMA region, the surface pressure sharply increases, reaching
elevated surface pressure values and highly compressed films.
Figure 4-9. Isotherms of the star-shaped PEO-b-PCL copolymers.
4.2.3.1 High MMA region
The first step toward understanding the behavior of the stars in this region was to check
monolayer formation reversibility and stability. This was done by carrying out compression-
expansion hysteresis experiments with target pressures up to 9 mN/m. For Star#6, Star#5, Star#4,
and Star#3, all the compression and expansion curves are superimposable independent of the
target pressure, which is indicative of film formation reproducibility and stability. These four
stars have the largest PCL amounts and are therefore hydrophobic enough so the amount of
74
material adsorbed at the interface remains constant over time, without irreversible dissolution in
the aqueous subphase (Figure 4-10). As the PCL content is decreased, monolayer stability is
reduced. For Star#2, the compression-expansion cycles start shifting toward smaller mean
molecular areas, and this shift is even more pronounced for Star#1 (Figure 4-11). For these two
samples, the small PCL chains are insufficient to overcome the overall star hydrophilicity arising
from the water-soluble PEO core, and irreversible water dissolution takes place over time.
Figure 4-10. Compression-expansion hysteresis plot of Star#6 (target pressure = 9 mN/m).
Figure 4-11. Compression-expansion hysteresis plot of Star#1 (target pressure = 9 mN/m).
75
From the initial study on the homopolymers, it was shown that in the low pressure region
ε-caprolactone repeat units are adsorbed at the A/W interface independent of the molecular
weight and that the PEO core alone has limited surface activity which leads to its irreversible
dissolution in the aqueous subphase. For the star-shaped block copolymers, the interfacial
behavior of the PEO core might be significantly changed as it is chemically attached to
hydrophobic PCL chains. To estimate the interfacial area occupied by the PEO core of the star-
shaped block copolymers, only the isotherms of Star#6, Star#5, Star#4, and Star#3 were used, as
the corresponding Langmuir monolayers are thermodynamically stable below the pseudoplateau.
For target pressures ranging from 1 to 11 mN/m, the MMA was plotted versus the number of ε-
caprolactone repeat units as shown in Figure 4-12. The resulting curves were analyzed by linear
regression leading to R2 values greater than 0.99. The linear relationships indicate that the
adsorption of the PCL blocks is not molecular weight-dependent as it is for linear PCL
homopolymers. More interestingly, the y-axis intercepts are significantly different from zero,
indicating the non-negligible interfacial area occupied by the PEO core anchored by the PCL
chains. The surface pressure was then plotted versus the y-axis intercept values to give the
extrapolated isotherm of the PEO core of the block copolymers (Figure 4-13). The experimental
PEO2000 isotherm was also included for comparison. Both curves have been normalized with
respect to the number of ethylene oxide units. For low surface pressures (π ≤ 4 mN/m), the
experimental PEO2000 and the extrapolated PEO core isotherms overlap reasonably, indicating a
similar interfacial area occupied by an ethylene oxide repeat unit of the PEO core compared to
the linear PEO analogue. The two curves stop overlapping above 4 mN/m because PEO2000 is
irreversibly dissolved in the aqueous subphase. It is also very interesting to notice in the
isotherms of the block copolymers, for surface pressures lower than 11 mN/m, the absence of
76
pseudoplateaus or inflection points characteristic of PEO aqueous dissolution. This absence is
also shown in the compressibility plot K versus surface pressure (Figure 4-14), where no peak
(local maximum) is observed below 11 mN/m (in compressibility plots versus surface pressure,
every phase transition in a Langmuir monolayer results in a local compressibility maximum).
This suggests that the PEO core of the star-shaped block copolymers is probably not adsorbed at
the interface but more likely already solvated in the water subphase in the vicinity of the
interface.
Figure 4-12. Plots of MMA versus number of ε-caprolactone repeat units for different surface pressures from the isotherms of Star#3, Star#4, Star#5, and Star#6.
Figure 4-13. Isotherms of the PEO core extrapolated and PEO2000 normalized with respect to the number of ethylene oxide units.
77
Figure 4-14. Compressibility plots of the star-shaped PEO-b-PCL block copolymers versus surface pressure.
4.2.3.2 Intermediate MMA region
This region is characterized by a pseudoplateau in the isotherms and by a maximum in the
compressibility plot (Figure 4-14). This phase transition occurs between approximately 12 and
15 mN/m, a similar pressure range as for the collapse pressure of PCL homopolymers. The
pseudoplateau length correlates with the PCL chain length in the stars, and this phase transition
is attributed to the PCL segments aggregating and crystallizing above the interface and the PEO
core, which is consistent with the work reported by Lee et al. on PCL-based block copolymers
with a dendritic hydrophilic head.120
Crystallization of the PCL segments was also characterized by carrying out isobaric
experiments for Star#6 with target pressures below and within the pseudoplateau pressure range
(9, 11, and 13 mN/m). The results of MMA decrease versus time are presented in Figure 4-15,
and the y-axis has been normalized to facilitate comparison. No or very little MMA decrease
takes place over time for pressures below the pseudoplateau (9 and 11 mN/m), which is
indicative of thermodynamically stable monolayers. As the target pressure is increased (13
78
mN/m), PCL quickly collapses and crystallizes as indicated by the sharp initial area decrease.
The MMA levels off around 1000 Å2, a value that correlates well with the MMA value obtained
at the end of the plateau in the isotherm.
Figure 4-15. Isobaric relaxation plots of Star#6 at 9, 11, and 13 mN/m.
PCL crystallization in the pseudoplateau region was finally investigated by hysteresis
experiments. The compression-expansion curves for Star#6 with a target pressure of 15 mN/m
are shown in Figure 4-16. As the barriers expand, an expansion pseudoplateau appears that
corresponds to the readsorption (melting) of the PCL chains previously crystallized.
Interestingly, the second and third compression curves overlap each other but do not overlap the
initial compression, with a slight shift toward the low MMA region and a decrease in the
compression pseudoplateau pressure. Similarly as it was observed for PCL homopolymers, this
confirms that crystallization takes place for Star#6 in the pseudoplateau region during
compression. Star#5, Star#4, and Star#3 have a similar behavior with the expansion
pseudoplateau vanishing as the PCL amount is decreased. Because of even lower PCL content in
Star#2 and Star#1, no expansion pseudoplateau was observed, and the subsequent compression-
79
expansion isotherms shifted toward smaller mean molecular areas because of irreversible
polymer dissolution in the water subphase (Figure 4-17), similarly as for low surface pressures.
Figure 4-16. Compression-expansion hysteresis plot of Star#6 (target pressure = 15 mN/m).
Figure 4-17. Compression-expansion hysteresis plot of Star#1 (target pressure = 15 mN/m).
4.2.3.3 Low MMA region
In the low MMA region, the surface pressure sharply increases up to values as high as 35
mN/m. Such high surface pressures can be reached because of the increased amphiphilicity of
the block copolymers compared to the homopolymers, even if the collapse pressure of the
individual PEO and PCL blocks is surpassed. The fact that the isotherms of Star#6, Star#5,
80
Star#4, and Star#3 overlap in this region suggests that all the PCL chains have collapsed and
crystallized above the interface, and that the sharp pressure increase arises mainly from
interactions between the hydrated PEO cores. Therefore, PCL crystallization probably takes
place for these four stars with the PEO core still hydrated in the water subphase, with the PCL
chains stretching and crystallizing away from the interface. The isotherms of Star#2 and Star#1
shift toward smaller mean molecular areas because of the water solubility behavior mentioned
before. A cartoon of the polymer chains’ conformations as the Langmuir monolayers are
compressed is proposed in Figure 4-18. It should be mentioned that the experiments discussed
here do not provide enough information to fully understand the aggregation of the star-shaped
PEO-b-PCL block copolymers. For instance, eventual phase separation between the different
blocks leading to the formation of various surface micelles could not be directly determined from
surface pressure measurements.
Figure 4-18. Proposed conformations modeling the adsorption of the star-shaped block copolymers at the A/W interface versus surface pressure.
81
4.2.3.4 AFM imaging
Typical AFM images of the star-shaped block copolymers LB films are presented in Figure
4-19 at pressures below (π = 10 mN/m) and above (π = 30 mN/m) the pseudoplateau. Similarly
as for PCL homopolymers, crystalline domains are observed for all the stars independent of the
surface pressure except for Star#1, which was unable to crystallize and only aggregated into
dewetted domains (~ 1-2 nm thick). Crystallization in PEO-b-PCL systems has been extensively
investigated in the previous literature, but the number of variables influencing crystallization
rates and morphologies is too large to allow any kind of generalization. For instance, Deng et al.
reported that, in PEO-b-PCL four-arm stars, PCL crystallinity for a constant PCL chain length
decreased as the PEO chain length was increased.114 Gan et al. observed that hydrophilic and
highly flexible PEO segments enhance the hydrophilicity and reduce the degree of crystallinity
of the polyester. The PEO block was also shown to provide nucleation sites for the crystallization
of the PCL block.137 Crystallization of PEO-b-PCL block copolymers in thin films resulted in
various types of spherulitic growths with crystallization of both PEO and PCL blocks.138-140
Nevertheless, it was widely demonstrated that PEO and PCL crystallize in well-defined
separated areas after their phase separation.141 Therefore, as PEO crystallization is obviously
difficult in LB films, probably as a result of residual hydration, we can reasonably assume that
the crystals obtained on mica substrates are the result of PCL crystallization only. Section
analysis indicates a constant crystal thickness around 7.5 nm, which is consistent with the
lamellae thickness obtained for PCL homopolymers.
82
Figure 4-19. Topographic AFM images of the star-shaped PEO-b-PCL copolymers LB films transferred below and above the pseudoplateau: Star#1 at 10 (a) and 30 mN/m (b); Star#2 at 10 (c) and 30 mN/m (d); Star#3 at 10 (e) and 30 mN/m (f); Star#4 at 10 (g) and 30 mN/m (h); Star#5 at 10 (i) and 30 mN/m (j); Star#6 at 10 (k) and 30 mN/m (l).
From the previously reported literature, the PCL crystal unit cell has a length of 1.7297 nm
that corresponds to two ε-caprolactone repeat units.142 PCL crystal thickness after transfer is
around 7.5 nm for all the star-shaped block copolymers, and if it is assumed that the PCL chains
orient perpendicularly to the mica substrate, this thickness indicates that the chains fold
approximately every eight ε-caprolactone repeat units. This would also support why the stars
with a number of ε-caprolactone repeat units per PCL chain greater or close to eight all
crystallized (Star#6, Star#5, Star#4, Star#3, and Star#2), whereas Star#1, with only three repeat
units per PCL chain, did not.
Similarly as for PCL homopolymers, the star-shaped samples exhibited PCL crystals at
low and high pressures, before and after crystallization of the PCL block at the A/W interface.
83
Additional crystallization of the PCL segments therefore also took place during and after transfer
for the star-shaped block copolymers, which likely leads to a mica surface only partially covered
with adsorbed or crystallized polymer chains. At high surface pressures, dendritic and needlelike
crystals could be seen. These crystal structures are significantly different from those observed for
PCL homopolymers, which indicates that even though the PEO core did not crystallize, it
strongly influenced the crystallization of the PCL blocks in terms of crystal morphology. Many
variables have to be taken into consideration, such as molecular weight, PEO amount, film
thickness, water evaporation, residual water content, transfer pressure, and substrate affinity for
the polymer chains to fully investigate crystallization of PCL homopolymers and PCL-based
block copolymers in LB films. While this study gives some hint of the surface properties of these
interesting star-shaped block copolymers, complete understanding of the system will therefore
require further investigation.
4.2.4 PEO-b-PCL Linear Diblock Copolymers
The PEO segment of the linear diblock copolymers (PEO2670, Mn = 2,670 g/mol) has a
molecular weight in the same range as the one of the PEO core of the star-shaped PEO-b-PCL
samples. Nevertheless, because of its linear architecture, this PEO segment has intrinsic surface
activity as shown in Figure 4-3, which should lead to the appearance of PEO phase transitions
not observable for the star-shaped samples in the low surface pressure region. The isotherms of
the four linear PEO-b-PCL diblock copolymers are presented in Figure 4-20. Compared to the
homopolymers PEO2670 and PCL2000 (PCL homopolymer with a molecular weight in the same
range as the PCL blocks of the linear diblock copolymers), higher surface pressures as high as 25
mN/m can be reached, similarly as for the star-shaped samples. As better shown in the
compressibility plot (Figure 4-21), three phase transitions that correspond to conformational
rearrangements of the polymer chains are clearly observed around 6.5, 10.5, and 13.5 mN/m. The
84
local maxima in monolayer compressibility for the two low pressure transitions increase as the
PCL chain length decreases, suggesting PEO-related phase transitions as previously observed for
comb-like polymers consisting of a poly(vinyl amine) backbone with 2kDa PEO side chains.143
The maximum in monolayer compressibility for the high pressure transition increases as the PCL
chain length increases, suggesting a PCL-related phase transition. Comparison with the
isotherms of PEO2670 and PCL2000 indicates that the transitions around 6.5 and 13.5 mN/m
arise from dissolution of the PEO block in the water subphase and crystallization of the PCL
block above the water surface, respectively. The transition at 10.5 mN/m was not observed for
PEO2670, but has been previously described as a brush formation of the PEO chains stretching
away from the interface when anchored by hydrophobic blocks for other PEO-based amphiphilic
block copolymers.144,145 In the following, we report our investigations on the behavior of the
linear PEO-b-PCL diblock copolymers in the low (π < 12 mN/m) and high (π > 12 mN/m)
surface pressure regions.
Figure 4-20. Isotherms of the PEO-b-PCL linear diblock copolymers.
85
Figure 4-21. Compressibility plots of the PEO-b-PCL linear diblock copolymers versus surface
pressure.
4.2.4.1 Low surface pressure region (π < 12 mN/m)
For surface pressures lower than the PEO aqueous dissolution around 6.5 mN/m, both the
PEO and the PCL blocks are adsorbed at the A/W interface in a pancake conformation. To
investigate the 2-dimensional miscibility of the two blocks in more details, we prepared
Langmuir monolayers with binary mixtures146-149 of PEO2670 and PCL2000, and the resulting
isotherms are presented in Figure 4-22. Contrary to the isotherms of the linear diblock
copolymers, only 2 phase transitions corresponding to PEO2670 aqueous dissolution (around 6.5
mN/m) and PCL2000 crystallization (around 13 mN/m) are observed. The absence of a phase
transition at 10.5 mN/m comes from the fact that PEO2670, which is not chemically anchored by
a hydrophobic PCL segment, is already irreversibly dissolved at this surface pressure in the
water subphase. This aqueous dissolution of PEO2670 (loss of material in the monolayer by
solubilization in the subphase) can be observed more clearly in the hysteresis experiment in
Figure 4-23 (49 mol % of PEO2670), with a target surface pressure of 9 mN/m, where the
86
successive compression-expansion curves shift toward the low MMA region only below 6.5
mN/m. Above 6.5 mN/m, the curves overlap because the amount of PCL2000 adsorbed at the
A/W interface stays constant, independent of the number of compression-expansion hysteresis
cycles.
Figure 4-22. Isotherms of PEO2670 and PCL2000 binary mixtures. (Inset) Corresponding compressibility plots versus surface pressure.
Figure 4-23. Compression-expansion hysteresis plot of the binary mixture with 49 mol % in PEO2670 (target pressure = 9 mN/m).
It is also interesting to notice that the collapse surface pressure of PEO2670 slightly
increases (up to 6.7 mN/m for 86 mol % of PCL2000) when increasing the amount of PCL2000
87
in the mixed monolayers, as shown by the shift of the compressibility maximum toward higher
surface pressures (inset of Figure 4-22). This is a first indication that PEO2670 and PCL2000
are miscible when both blocks are adsorbed at the A/W interface.150 Figure 4-24 shows plots of
MMA versus the mole fraction of PCL2000 for three surface pressures below 6.5 mN/m (2, 3,
and 4 mN/m). The data exhibit negative deviations from ideal mixing (dashed lines), which
confirms that in the surface pressure range where they are adsorbed at the A/W interface (π < 6.5
mN/m), PCL2000 and PEO2670 do not phase-separate and thermodynamically interpenetrate
each other.151,152 From these results, we can reasonably extrapolate that the PEO and PCL
segments of the linear diblock copolymers are miscible as well below 6.5 mN/m, and therefore
that their A/W interfacial adsorption probably does not lead to the formation of surface micelles
previously observed for other amphiphilic block copolymers.97,153,154 These results are in good
agreement with a previous study that demonstrated the miscibility of PEO and PCL
homopolymers in the amorphous phase in blend films prepared by solution casting.155
Figure 4-24. MMA plots versus mole fraction of PCL2000. Dashed lines: theoretical ideal mixing.
88
The reversibility of the two PEO phase transitions of the linear diblock copolymers was
investigated by carrying out compression-expansion hysteresis experiments110,156 on the block
copolymer sample with the smaller PCL segment (PEO60-b-PCL11) for a target pressure of 18
mN/m. The resulting π/MMA curves and the corresponding compressibility plots are shown in
Figures 4-25 and 4-26, respectively. After the first compression, the curves in Figure 4-25 shift
to the low MMA region, which is indicative of some irreversibility in the PEO phase transitions.
As shown in the compressibility plots, the local maximum at 6.5 mN/m disappears after the first
compression whereas the maximum at 10.5 mN/m is still present, independent of the number of
compression-expansion cycles. At 6.5 mN/m, we believe the PEO chains dissolve irreversibly in
the aqueous subphase and adopt a mushroom conformation. Nevertheless, because of the
anchoring effect of the PCL segments, the PEO chains stay in the vicinity of the interface. Upon
further monolayer compression, the PEO chains are compressed against each other and stretch
perpendicularly to the interface to form a compact brush at 10.5 mN/m.22,23,144,157-160 During
monolayer expansion, the PEO brush reversibly relaxes, but the PEO chains do not readsorb and
stay hydrated underneath the interface, which explains the complete absence of phase transition
at 6.5 mN/m after the first compression. The maximum in compressibility corresponding to the
transition at 10.5 mN/m is nevertheless slightly decreased after the first compression, which
suggests that the PEO chains do not completely relax during the subsequent expansions to their
original mushroom conformation of the first compression. It should be noticed that, contrary to
our linear PEO-b-PCL diblock copolymers, only one apparent PEO phase transition around 10
mN/m is usually observed in the isotherms of PEO-based block copolymers with high molecular
weight PEO blocks (Mn ≥ 10,000 g/mol),22 probably because PEO aqueous dissolution and brush
formation take place simultaneously.
89
Figure 4-25. Compression-expansion hysteresis plot of PEO60-b-PCL11 (target pressure = 18 mN/m).
Figure 4-26. Compressibility plots of Figure 4-25 (PEO60-b-PCL11, target pressure = 18 mN/m). Top curve: first compression. Bottom curve: first expansion, second compression, and second expansion.
90
4.2.4.2 High surface pressure region (π > 12mN/m)
In this surface pressure region, similarly as for the PCL homopolymers and the star-shaped
block copolymers, the phase transition observed in Figures 4-20 and 4-21 at 13.5 mN/m arises
from collapse and crystallization of the PCL segments above the water surface. The
compression-expansion hysteresis experiment carried out on the linear block copolymer sample
with the longest PCL block (PEO60-b-PCL35) for a target pressure of 16 mN/m gives more
insight on the crystallization/melting behavior of the PCL segments. The π/MMA curves and the
corresponding compressibility plots are shown in Figures 4-27 and 4-28, respectively. During
monolayer compression, the hydrophobic PCL segments collapse and crystallize on top of the
water surface as shown by the inflection points in the π/MMA curves and the local maxima in
the compressibility plots at 12.5 and 13.5 mN/m. Similarly as for the PCL homopolymers and for
the star-shaped samples, the collapse/crystallization surface pressure in the first cycle is
approximately 1 mN/m higher than in the subsequent compressions (13.5 versus 12.5 mN/m),
and also, after the first compression, the subsequent curves are slightly shifted toward the low
mean molecular area region. After the first hysteresis cycle, the PCL crystals that did not melt
(i.e. did not readsorb at the A/W interface) act as nucleation sites to catalyze the crystallization of
other PCL segments during the subsequent cycles, therefore lowering the crystallization surface
pressure and shifting the compression isotherms toward lower mean molecular areas. During
monolayer expansion, melting of the PCL segments was characterized in the π/MMA plots for
the star-shaped samples and the PCL homopolymers by an expansion pseudoplateau at a surface
pressure lower than for the crystallization. As shown in Figures 4-27 and 4-28, no pseudoplateau
or sharp local maximum in compressibility are observed during the first expansion, whereas for
the subsequent cycles the PCL segments clearly readsorb around 5 mN/m (sharp local maximum
91
in the compressibility plot). The presence of a broad melting transition during the first monolayer
expansion is a little bit surprising because this behavior was not previously observed for the star-
shaped samples and the PCL homopolymers, but it is probably related to a particular shape, size,
or polydispersity of the PCL crystals formed during the first monolayer compression. It should
be noticed that the crystallization and melting surface pressures reported here for linear diblock
copolymers (as well as for the star-shaped block copolymers and the PCL homopolymers) are
not thermodynamic values and were shown to be strongly barrier speed dependent. If one could
compress and expand the monolayers infinitely slowly, the crystallization and the melting
surface pressures would respectively decrease and increase, probably leveling off to a common
value.
Figure 4-27. Compression-expansion hysteresis plot of PEO60-b-PCL35 (target pressure = 16 mN/m).
92
Figure 4-28. Compressibility plots of Figure 4-27 (PEO60-b-PCL35, target pressure = 16 mN/m). a: PCL crystallization at 13.5 mN/m during the 1st compression. b: PCL crystallization at 12.5 mN/m during the 2nd and 3rd compressions. c: broad PCL melting transition during the 1st expansion. d: PCL melting transitions during the 2nd and 3rd expansions.
Crystallization of the linear PEO-b-PCL diblock copolymers in the LB monolayers was
finally evidenced by AFM imaging after transfer onto mica substrates for surface pressures
above 13.5 mN/m, after crystallization of the PCL segments on the water surface took place. As
shown in Figure 4-29 for a surface pressure of 15 mN/m, all the samples show hair-
like/needlelike crystal structures, with a constant crystal thickness around 7.5 nm as determined
from cross-section analysis. This thickness is consistent with the value obtained for the PCL
homopolymers and the star-shaped block copolymer samples. The PEO chains are adsorbed onto
the mica substrate and the PCL segments stretch perpendicularly to the interface, folding
approximately every 8 ε-caprolactone repeat units. These results confirm what we observed for
the star-shaped PEO-b-PCL samples, where the PEO block highly influenced crystallization of
the PCL segments in terms of crystal morphology. Nevertheless, it is crucial to emphasize here
93
once again that this brief crystal structure analysis is done for the LB films only and is not
rigorously valid for Langmuir monolayers, because film drying during LB film formation can
lead to further crystallization of the PCL segments. Evidence for crystallization of the PCL
segments in the star-shaped and the linear block copolymer samples directly at the A/W interface
came from the isotherms, the hysteresis, and the isobaric experiments, but no clear conclusions
can yet be drawn in terms of crystal structures in the Langmuir monolayers. BAM experiments
are currently underway to investigate more deeply the in-situ and real-time PCL crystal
growth/melting directly on the water surface for the star-shaped and the linear PEO-b-PCL block
copolymers.
Figure 4-29. Topographic AFM images of the linear PEO-b-PCL diblock copolymers LB films transferred after crystallization of the PCL segment at the A/W interface (π = 15 mN/m). (a): PEO60-b-PCL11 (b): PEO60-b-PCL19 (c): PEO60-b-PCL27 (d) and (e): PEO60-b-PCL35 (f): Cross-section analysis on PEO60-b-PCL35
94
A cartoon summarizing the polymer chains conformations as the Langmuir monolayers of
the linear PEO-b-PCL diblock copolymers are compressed is proposed in Figure 4-30. For
surface pressure values lower than 6.5 mN/m, both the PEO and the PCL segments are adsorbed
at the A/W interface in a pancake conformation (1). As the surface pressure is increased, the
PEO segments irreversibly dissolve in the aqueous subphase and adopt a mushroom
conformation around 6.5 mN/m (2), before forming a brush above 10.5 mN/m upon further
compression (3). Finally, around 13.5 mN/m, the PCL segments collapse and crystallize
perpendicularly to the interface (4).
Figure 4-30. Proposed conformations modeling the adsorption of the linear PEO-b-PCL diblock copolymers at the A/W interface versus surface pressure.
95
4.3 Conclusions
In this chapter, the A/W interfacial behavior of various PEO-b-PCL block copolymers and
their LB film morphologies on hydrophilic mica substrates were investigated, and the results
were compared to PEO and PCL homopolymers. The isotherms of the star-shaped block
copolymers indicated the presence of a single phase transition characterized by a pseudoplateau
that corresponds to the collapse and crystallization of the PCL chains above the water surface.
Below the plateau, the PCL segments are adsorbed, anchoring the water-soluble star-shaped PEO
core in the vicinity of the interface. Compression-expansion hysteresis experiments showed that,
in this region, the spread monolayers are thermodynamically stable except the ones containing
the smallest PCL amounts, which irreversibly dissolved in the water subphase. In the
pseudoplateau region, PCL homopolymers crystallized directly at the A/W interface as well as
the PCL segments of the star-shaped block copolymers. Above the pseudoplateau, the isotherms
of the star-shaped block copolymers with the longest PCL chains overlapped, indicating that all
the PCL chains have collapsed and that the sharp pressure increase mainly arises from
interactions between the hydrated PEO cores. Compression-expansion hysteresis experiments
indicated that the readsorption/melting of the PCL segments takes place at a lower surface
pressure than for the crystallization. AFM imaging of the homopolymers and the star-shaped
block copolymers LB films was complicated by the fact that both PEO and PCL are highly
crystalline polymers that can undergo morphological changes during monolayer transfer. The
PEO homopolymers did not crystallize, probably because residual hydration or large hydrophilic
substrate/PEO monolayer interactions inhibited crystal formation. The PCL homopolymers and
the star-shaped block copolymers crystallized directly at the A/W interface only above the PCL
collapse pressure, but additional crystallization could take place during water evaporation on the
mica substrates. Various crystal morphologies were observed for the star-shaped block
96
copolymers such as spherulitic, dendritic, and needlelike structures, with the presence of the PEO
core strongly influencing the crystallization of the PCL blocks. The linear diblock copolymers
successfully self-assembled as well at the A/W interface to form stable Langmuir monolayers.
Preliminary investigation on PEO2670 and PCL2000 homopolymers blends showed that these
polymers are non-ideally miscible for low surface pressures when both blocks are adsorbed at the
A/W interface. Nevertheless, the individual collapse surface pressures (PEO2670 aqueous
dissolution around 6.5 mN/m and PCL2000 crystallization above the interface around 13 mN/m)
were not significantly influenced by the presence of the other homopolymer. For the linear PEO-
b-PCL diblock copolymers, an additional PEO phase transition at 10.5 mN/m was observed
corresponding to the formation of a PEO brush underneath the anchoring PCL segments. These
two PEO phase transitions were not observed for the star-shaped PEO-b-PCL block copolymers,
and our investigations consequently confirmed the significant influence of the polymer
architecture on its interfacial properties. AFM imaging of the linear PEO-b-PCL diblock
copolymers LB films for high surface pressures confirmed the formation of PCL crystals with
hairlike/needlelike architectures. These crystals were significantly different from those obtained
in LB films of PCL homopolymers, confirming the strong influence of the PEO block on the
crystallization of the PCL segments. This fundamental investigation gave interesting insight on
the interfacial self-assembly of PEO-b-PCL copolymers and showed that an accurate and easy
control of the conformations and the orientations of the different blocks at the A/W interface can
be easily achieved by simply varying the polymer architecture or the surface pressure.
4.4 Experimental Methods
4.4.1 Langmuir Films
Surface pressure measurements were accomplished by use of a Teflon Langmuir trough
system (W = 160 mm, L = 650 mm; KSV Ltd., Finland) equipped with two moving barriers and a
97
Wilhelmy plate. Between runs, the trough was cleaned with ethanol and rinsed several times with
Millipore filtered water (resistivity ≥ 18.2 MΩ.cm). The samples were typically prepared by
dissolving approximately 1 mg of polymer in 1 mL of chloroform. Volumes ranging from 10 to
30 µL were spread dropwise on a Millipore filtered water subphase with a gastight Hamilton
syringe. The chloroform was allowed to evaporate for 30 min to ensure no residual solvent
remained. When not in use, the volumetric flasks containing the polymer solutions were wrapped
with Teflon tape followed by Parafilm and stored at 10 °C in order to prevent changes in
concentration due to chloroform evaporation. In all the experiments, subphase temperature and
barrier speed were kept constant at 25 °C and 5 mm/min, respectively, unless otherwise stated.
4.4.2 AFM Imaging
The LB films were formed by transferring the Langmuir films of the PEO-b-PCL block
copolymers (linear and star-shaped) and the homopolymers onto freshly cleaved mica at the
desired surface pressure which was attained at compression/expansion rates of +/-5 mm/min.
Once the films had equilibrated at a constant surface pressure for 15 min, the mica substrate was
then pulled out of the water subphase at a rate of 1 mm/min. All the transfer ratios were close to
unity unless otherwise stated, which is indicative of successful transfer. The transferred films
were air-dried in a desiccator for 24 h and subsequently scanned in tapping mode with a
Nanoscope III AFM (Digital Instruments, Inc., Santa Barbara, CA) by use of Nanosensors
silicon probes (dimensions: T = 3.8-4.5 µm, W = 26-27 µm, L = 128 µm). All the images were
processed with a second-order flattening routine.
98
CHAPTER 5 TWO-DIMENSIONAL POLYMERIC NANOMATERIALS THROUGH CROSS-LINKING
OF POLYBUTADIENE-b-POLY(ETHYLENE OXIDE) MONOLAYERS AT THE AIR/WATER INTERFACE
5.1 Introduction
The idea of stabilizing amphiphilic self-assemblies by polymerization was introduced at
least thirty years ago for monolayers and about ten years later for bilayer vesicles.161,162 This
approach to bridging the nanoscale world of labile, interfacially driven self-assemblies with the
meso-scale has resulted in several examples of cross-linked 3D structures.163-167 For example,
Bates and co-workers were the first to succeed in retaining the cylindrical morphology formed by
gigantic wormlike rubber micelles of polybutadiene-b-poly(ethylene oxide) (PB-b-PEO) diblock
copolymers in water by chemical cross-linking of the PB cores through their pendant 1,2-double
bonds.163,167,168 However, relatively few groups have shown interest in stabilization by cross-
linking of two-dimensional (2D) polymeric self-assemblies formed at the air/water (A/W)
interface; most studies have involved interfacial polymerization of small molecules in Langmuir
monolayers.169-198
In the early 1970’s, Veyssié and co-workers180,190,191,193 were the first to demonstrate the
formation of 2D cross-linked materials by cross-linking monolayers of dimethacrylates and
several other difunctional reactive amphiphiles under UV irradiation for a constant surface
pressure at the A/W or the oil/water interface. This idea inspired other research groups and
several examples followed. Regen and co-workers introduced the concept of a 2D-network of
molecular pores, i.e. “perforated monolayers” derived from calix[n]arene-based amphiphiles.183-
188 Cross-linking with malonic acid or via UV irradiation enabled them to synthesize porous and
cohesive “perforated monolayers” with pore diameters in the range 2-6 Å potentially applicable
for gas permeation selectivity.186,187,189 Michl and co-workers synthesized grids through the
99
coupling of star-shaped monomers forced to adhere to a mercury surface.194,195 After
polymerization, well-defined covalent 2D square- or hexagonal-grid polymers could be
synthesized,194,195,199 and analogous supramolecular routes were also proposed.200-202 Palacin and
co-workers reported on cross-linking porphyrins through molecular recognition between
oppositely charged monomers at the A/W interface.196-198 Kloeppner and Duran171 were the first
to demonstrate the possibility to remove from the water surface free-standing fibers of 2D cross-
linked 1,22-bis(2-aminophenyl)docosane polyanilines. Finally, alkylalkoxysilanes have been
widely used,177-179,203,204 and our group has for instance investigated some of the fundamental
aspects of the A/W interfacial cross-linking of octadecyltrimethoxysilane (OTMS) and
octadecyltriethoxysilane (OTES) under acidic conditions.177-179,203 However, relatively few
groups have shown interest in stabilizing by cross-linking 2D “true” polymeric self-assemblies at
the A/W interface. To our knowledge, only one example based on a lipopolymer was previously
proposed by O’Brien and co-workers involving network formation by photopolymerization.205
Our interest is to cross-link monolayers of block copolymers to achieve porosity at the sub-
micrometer scale. In this chapter, the synthesis of a 2D polymeric nanomaterial consisting of a
continuously cross-linked PB network containing PEO domains of controllable size is illustrated.
This work was done in collaboration with Rachid Matmour, graduate student in the Duran group
at the University of Florida. Such thin films have potential applications in the preparation of
membranes which will show large differences in permeability to water, methanol, and other
polar compounds, depending on the PEO “pore” size.
We report in this chapter the 2D self-condensation, at the A/W interface and under acidic
conditions, of a triethoxysilane-functionalized PB-b-PEO three-arm star block copolymer (PB
core and PEO corona) and of a triethoxysilane-functionalized linear PB homopolymer in a
100
preliminary investigation. Using PB, which pendant double bonds have been hydrosilylated with
trialkoxysilanes, as the cross-linkable block is a novel approach that can be applicable to a
variety of other polydiene-based block copolymers in order to retain a specific morphology at the
nanoscopic scale. The surface properties of the cross-linked monolayers were characterized by
surface pressure measurements such as surface pressure (π)-mean molecular area (MMA)
isotherms at different reaction times, and isobaric experiments for various subphase pH values.
The morphologies of the Langmuir monolayers were studied by atomic force microscopy (AFM)
imaging of the corresponding Langmuir-Blodgett (LB) films.
5.2 Results and Discussion
5.2.1 Hydrosilylated PB Homopolymer
5.2.1.1 Hydrosilylation reaction
To demonstrate the viability of the 2D cross-linking method, we chose to first focus on a
commercially available linear PB homopolymer (Mn = 11,050 g/mol, ~ 204 butadiene repeat
units). Many publications and patents can be found in the literature on the hydrosilylation of
polymers.206-217 In most cases, the hydrosilylated polydienes were used as precursors to
synthesize macromolecular complex architectures such as arborescent graft polybutadienes,218
multigraft copolymers of PB and polystyrene,219 or side-loop polybutadienes.220 Triethoxysilane
was used here as the pendant double bond hydrosilylating agent in stoichiometric amount with
the total molar amount of repeat units in the PB homopolymer and in the presence of Karstedt
catalyst (platinum catalyst) as shown in Figure 5-1. The reaction was carried out under argon for
24 h at 80 °C in dry toluene (water free environment). After workup, the hydrosilylated PB was
analyzed by 1H NMR and FTIR spectroscopies (Figures 5-2, 5-3, and 5-4).
101
Figure 5-1. Hydrosilylation of the pendant double bonds of the PB homopolymer.
Figure 5-2. 1H NMR spectrum of the PB homopolymer.
102
Figure 5-3. 1H NMR spectrum of the hydrosilylated PB homopolymer.
Figure 5-4. FTIR spectra of the PB homopolymer before and after hydrosilylation.
The 1H NMR spectrum of the PB starting material was used to determine the distribution
of 1,2- and 1,4-units. The two protons of the pendant vinyl carbon in the 1,2-units (=CH2) and
the other hydrogens in the double bonds (-CH=CH- and –CH=CH2) having chemical shifts of 4.9
and 5.4 ppm, respectively, the PB homopolymer turned out to be composed of 89 mole % of 1,2-
103
units (Figure 5-2). The 1H NMR spectrum of the hydrosilylated PB revealed a strong decrease in
the intensity of the signal corresponding to the -CH=CH2 (δ = 4.9 ppm) protons. Furthermore,
the appearance of intense peaks at δ = 1.2 and 3.8 ppm corresponding respectively to the –Si–
OCH2CH3 methyl protons and the –Si–OCH2CH3 methylene protons is indicative of a high
degree of conversion. However, some pendant double bonds remained unreacted after
hydrosilylation (Figure 5-3). Based on the integration values of the signals at δ = 4.9 and 5.4
ppm, a conversion of 75 % of the 1,2-PB pendant double bonds was found, assuming that
triethoxysilane reacts predominantly with the 1,2-PB units as previously demonstrated.221 This
result was confirmed by FTIR spectroscopy as shown in Figure 5-4, where the absorbance peaks
at 3100 cm-1 (=CH2 anti-symmetric stretch) and 1640 cm-1 (alkenyl –HC=CH2 stretch) strongly
decreased in intensity after hydrosilylation.
5.2.1.2 Cross-linking reaction at the A/W interface
After characterization of the PB68-co-PB(Si(OEt)3)136 triethoxysilane-functionalized PB, its
A/W interfacial cross-linking by self-condensation of the triethoxysilane groups was studied.
This 2D acid-catalyzed condensation reaction involves two different steps as shown in Figure 5-
5: hydrolysis of the ethoxy groups with elimination of ethanol, followed by condensation
between the resulting silanols. Several isotherms were first recorded after different reaction times
(subphase pH = 3.0) as shown in Figure 5-6 with a fast barrier compression speed (100 mm/min)
to prevent additional cross-linking during monolayer compression. As the reaction time is
increased, the isotherms shift toward the low mean molecular area region because of the
irreversible loss of ethanol and water molecules into the water subphase during the hydrolysis
and condensation steps, respectively. As shown in Figure 5-7, the monolayer’s static elastic
modulus εs (calculated from Equation 3-1101) significantly increases versus reaction time,
104
indicating that the material becomes more and more rigid as the extent of cross-linking is
increased. For reaction times longer than 10 h, the isotherms essentially overlapped which
indicates completion of the cross-linking.
Figure 5-5. Cross-linking reaction involving hydrolysis and condensation of the triethoxysilane groups.
Figure 5-6. Surface pressure-Mean Molecular Area isotherms of the hydrosilylated PB carried out after different reaction times (subphase pH = 3.0).
105
Figure 5-7. Static elastic modulus-surface pressure curves of the hydrosilylated PB homopolymer at different reaction times (subphase pH = 3.0).
From the isotherms, the interfacial area occupied by one silane repeat unit before reaction
and its decrease during cross-linking were estimated (π = 5 mN/m, pH = 3.0) and compared with
the values previously reported for OTES under similar experimental conditions. The MMA for
the hydrosilylated PB decreases from 6300 Å2 (46 Å2/silane repeat unit) down to 3520 Å2 (26
Å2/silane repeat unit), which corresponds to a decrease (∆A) of approximately 20 Å2/silane
repeat unit. These values are in very good agreement with the ones reported for OTES (46
Å2/molecule before cross-linking, 24 Å2/molecule after cross-linking, and ∆A = 22 Å2/molecule)
and clearly indicate that the extent of Sol-Gel cross-linking is not reduced when starting from
true polymeric chains compared to single alkylalkoxysilane molecules.
The pH influence on the cross-linking reaction kinetics was shown by carrying out isobaric
experiments at π = 10 mN/m and for different subphase pH values as shown in Figure 5-8. As
expected, the MMA decreases faster for lower pH values. The isobar at pH = 7.0 shows a very
106
slow creep over time which demonstrates that the reaction is likely insignificant under neutral pH
conditions. For lower pH values (pH = 2.0 and 3.0), the curves overlap with the MMA leveling
off after about 7 h indicating completion of the cross-linking reaction. The cross-linking kinetics
are consistent with those reported for OTES and are slower compared to the results obtained for
OTMS,177-179,203 which is related to the slower elimination of larger alkoxy substituents during
the hydrolysis step.
Figure 5-8. MMA-time isobars of the hydrosilylated PB for various subphase pH values (π = 10
mN/m).
Upon completion of the cross-linking reaction, the cross-linked material could be
subsequently manually removed from the interface with a spatula after its compression to a final
area of ca. 2 x 15 cm2 (Figure 5-9), leading to a film approximately 50 monolayers thick. It was
self-supporting and gel-like, and could be collected as elongated sheets, which in turn could be
drawn into very long fibers at high elongation. As expected, it was insoluble in common organic
solvents such as chloroform or THF, making molecular weight analysis by SEC impossible.
107
Figure 5-9. (A and B) Removal of the cross-linked hydrosilylated PB from the water surface. (C) The long cross-linked vacuum-dried fiber.
5.2.1.3 AFM imaging
The evolution of the monolayer morphology during cross-linking was characterized by
AFM imaging of the LB films transferred onto mica substrates (Figure 5-10, π = 10 mN/m). As a
control experiment, it was first observed that under neutral pH conditions (pH = 7.0, no cross-
linking), the hydrosilylated PB forms a smooth and featureless monolayer (Figure 5-10B), in
opposition to the highly hydrophobic PB starting material which forms typical rubbery
continuous aggregates above the water surface (Figure 5-10A). After its hydrosilylation, the PB
becomes amphiphilic (hydrophobic backbone and hydrophilic triethoxysilane side groups) and
consequently surface active with the triethoxysilane pendant groups solvated into the water
subphase. This interfacial property of the hydrosilylated PB was also shown in the isotherms
108
where stable monolayers could be formed for surface pressures as high as 40 mN/m before
collapsing (Figure 5-6). After 20 minutes of reaction (≈ 50 % extent of cross-linking according to
the isobar at pH = 3.0 and π = 10 mN/m), the cross-linked material becomes more hydrophobic
and can be clearly observed in Figure 5-10C (bright areas) with an average height of 1 nm as
determined by cross-section analysis (Figure 5-10E). The cross-linked PB has irregular borders
and does not cover yet the entire mica surface. An AFM image obtained after completion of the
cross-linking reaction is shown in Figure 5-10D (10 h, pH = 3.0, π = 10 mN/m). Under these
experimental conditions, most of the mica surface was covered with a smooth and cross-linked
monolayer. Therefore, we deliberately found an area with a crack (that probably formed during
film transfer) to clearly show the presence of the cross-linked monolayer (bright area) on top of
the mica substrate with a thickness that stays constant around 1 nm during cross-linking (Figure
5-10F).
The acid-catalyzed condensation between the triethoxysilane pendant groups of a
hydrosilylated PB obtained by hydrosilylation of a commercial PB homopolymer has been
successfully applied in this preliminary investigation to the preparation of cross-linked polymeric
monolayers without any reagents or additives directly at the A/W interface. This technique opens
up the possibility to retain a specific 2D morphology at the nanoscopic scale as exemplified in
the following part for PB-b-PEO three-arm star block copolymers.
109
Figure 5-10. AFM topographic images of the LB films transferred onto mica substrates at π = 10 mN/m: the commercial PB homopolymer (A) and the corresponding hydrosilylated PB at pH = 7.0 (B; t = 0 h) and 3.0 for different reaction times (C ; t = 20 min and D; t = 10 h). (E and F) Cross-section analysis of the images C and D. The images are 7 x 7 µm2 (A) and 50 x 50 µm2 (B, C, and D).
5.2.2 Hydrosilylated PB-b-PEO Three-Arm Stars
The surface properties of a new set of (PB-b-PEO)n (n = 3 or 4) amphiphilic three- and
four-arm star block copolymers at the A/W interface were recently investigated in our group.222
A divergent anionic polymerization method yielded star-shaped block copolymers with well-
110
defined architectures, molecular weights, and block volume fractions. Different (PB-b-PEO)3
amphiphilic three-arm star block copolymers exhibiting narrow molecular weight distributions
were prepared with poly(ethylene oxide) coronas over a broad range of volume fractions as
shown in Table 5-1. Isothermal characterization of the three-arm stars at the A/W interface
indicated the presence of three characteristic regions: a “pancake” region (I) in the high MMA
region where the surface pressure slowly increases as the monolayer is compressed; a
pseudoplateau around 10 mN/m (II) that corresponds to the aqueous dissolution of the PEO
chains; and finally a compact brush region (III) in the low MMA region where the sharp increase
in surface pressure originates only from the interactions between the hydrophobic PB segments
(Figure 5-11). The dotted lines in Figure 5-12 also show the extrapolations used to estimate the
three corresponding parameters Apancake, Ao, and ∆A. A fit of the pseudoplateau data revealed a
linear dependence of ∆A with the number of EO units (y = 12.351x - 0,4889; R2 = 0.99),
indicating that the length of the pseudoplateau linearly increases as the amount of PEO in the
star-shaped block copolymers is increased (Figure 5-13).
Table 5-1. Number average molecular weights and polydispersity indexes of the three-arm star block copolymers.
Run Mna
(SEC) Mn
b (1H NMR)
Mnc
(theo.) Mw/Mna) Code
1 45900 42500 40500 1.2 (PB200-b-PEO76)3
2 56000 75500 77500 1.15 (PB200-b-PEO326)3
3 58000 160500 164500 1.2 (PB200-b-PEO970)3
4 74000 320500 323000 1.2 (PB200-b-PEO2182)3
5 - 135500 - - (PB(Si(OEt)3)-b-PEO)3 a Apparent molecular weights determined by SEC in THF using a polystyrene calibration. b Estimated by 1H NMR analysis. c Mn,th = MButadiene x ([Butadiene]/[-PhLi]) x 3 + MEO x ([EO]/[(PB-OH)3].
111
Figure 5-11. Surface Pressure-MMA isotherms for the (PB200-b-PEOn)3 three-arm star block copolymers (n = 76, 326, 970, and 2182).
Figure 5-12. Isotherm of (PB76-b-PEO444)4 depicting how Apancake, Ao, and ∆Apseudoplateau are determined.
112
Figure 5-13. Linear dependence of ∆Apseudoplateau on the total number of ethylene oxide units.
5.2.2.1 Hydrosilylation reaction
Similarly as for the PB homopolymer, the hydrosilylation reaction was applied here to the
PB segments of the (PB200-b-PEO326)3 three-arm star block copolymer using triethoxysilane in
stoichiometric amount with the total molar amount of double bonds in the PB block (1,2 and 1,4
units) as shown in Figure 5-14. After workup, the hydrosilylated block copolymer was analyzed
by 1H NMR and FTIR spectroscopies (Figures 5-15 and 5-16).
Figure 5-15 shows the 1H NMR spectra of the (PB-b-PEO)3 three-arm star block
copolymer before and after hydrosilylation. The 1H NMR spectrum of the starting material was
used to determine the distribution of 1,2- and 1,4-units in the PB block, similarly as previously
113
discussed for the PB homopolymer. The PB block of the star-shaped block copolymer turned out
to be composed, before hydrosilylation, of 75 mole % of 1,2-PB units. The 1H NMR spectrum of
the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 three-arm star block copolymer revealed a strong
decrease in the intensity of the signal corresponding to the -CH=CH2 protons of the pendant
double bonds at δ = 4.9 ppm. Furthermore, the fact that the signal corresponding to the
SiOCH2CH3 methyl protons at δ = 1.2 ppm greatly increased in intensity indicated that the
reaction occurred with a high efficiency, and a conversion of 85 % of the 1,2-PB pendant double
bonds was calculated. The efficiency of the reaction was confirmed by FTIR spectroscopy as
shown in Figure 5-16, where the absorbance peaks at 3100 cm-1 and 1640 cm-1 strongly
decreased in intensity after hydrosilylation.
Figure 5-14. Hydrosilylation of the pendant double bonds of the (PB-b-PEO)3 three-arm star block copolymers.
114
Figure 5-15. 1H NMR spectra of the (PB200-b-PEO326)3 star block copolymer and the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer.
Figure 5-16. FTIR spectra of the (PB200-b-PEO326)3 star block copolymer and the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer.
115
5.2.2.2 Cross-linking reaction at the A/W interface
The A/W interfacial behavior of the (PB(Si(OEt)3)-b-PEO)3 three-arm star block
copolymer was first studied through isotherm experiments for a subphase pH value of 3.0
(Figure 5-17). The (PB(Si(OEt)3)-b-PEO)3 star was initially spread on the water surface, the
cross-linking reaction (pH = 3.0) was carried out for 10 h in the liquid expanded region of the
isotherm at zero pressure, and then the isotherm was recorded as shown in Figure 5-17 (bottom
red curve). For comparison, the top blue curve illustrates the same sample spread and rapidly
compressed (barrier compression speed = 100 mm/min) at pH = 3 before any significant
triethoxysilane hydrolysis or condensation occurs. The isotherm of the non-hydrosilylated star is
also included (middle black curve). The (PB(Si(OEt)3)-b-PEO)3 star block copolymer before
cross-linking occupies a larger interfacial area compared to the non-hydrosilylated star because
of the molecular weight increase during reaction and also because of the increased affinity for
the A/W interface of the hydrosilylated PB block that spreads better than the highly hydrophobic
non-hydrosilylated PB block. Concerning the isotherm recorded after completion of the cross-
linking reaction (bottom red curve), a significant shift toward the low mean molecular area
region was observed, similarly as for the hydrosilylated PB homopolymer, because of the loss of
ethanol and water molecules. Another interesting feature is that the pseudoplateau at 10 mN/m
almost completely vanished for the unreacted hydrosilylated material, while it reappears upon
cross-linking. This pseudoplateau corresponds to the desorption of the PEO chains from the
interface into the aqueous subphase.223,224 Since the interfacial area occupied by the PB blocks
after their hydrosilylation significantly increases, the fractional area occupied by the PEO
segments therefore significantly decreases (even though the total area occupied by the PEO
segments stays constant), which leads to a PEO phase transition (pseudoplateau) much less
pronounced. As the cross-linking reaction proceeds, the area occupied by the cross-linked PB
116
segments decreases, which results in an increase of the fractional area occupied by the PEO
segments and, as a consequence, in a more pronounced pseudoplateau.
Figure 5-17. Surface Pressure-MMA isotherms of the (PB200-b-PEO326)3 star block copolymer
and of the corresponding hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer before and after cross-linking.
Similarly as for the hydrosilylated PB homopolymer, several isotherms were also
recorded for the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer after different
reaction times (pH = 3.0) as shown in Figure 5-18A. Fresh monolayers were spread for every
isotherm and the barrier compression speed was set to 100 mm/min to prevent additional cross-
linking during compression. As the reaction proceeds, the isotherms shift toward the low MMA
region because of a more compact cross-linked material, with the PEO pseudoplateau becoming
more and more pronounced because of the PEO fractional area influence mentioned above. The
reappearance of the PEO pseudoplateau is better shown by plotting monolayer compressibility
versus MMA for different reaction times as shown in Figure 5-18B, where the maximum in
117
compressibility, which is indicative of the PEO phase transition, significantly increases as the
cross-linking proceeds. The monolayer compressibility (K) was calculated from Equation 3-2.
Figure 5-18. Surface Pressure-MMA isotherms (A) and compressibility-MMA curves (B) of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer at various reaction times (subphase pH = 3.0).
118
The pH influence on the cross-linking reaction kinetics is shown by the isobaric
experiments carried out at π = 5 mN/m for different subphase pH values. This low surface
pressure was chosen to avoid the region of the PEO-related phase transition (pseudoplateau),
where the brush formation could lead to an additional decrease in MMA. As shown in Figure 5-
19, the results are consistent with those obtained for the hydrosilylated PB homopolymer. As
expected, the MMA decreases faster for lower pH values, and the MMA levels off after about 7
h for pH = 3.0.
Figure 5-19. Isobars of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star block copolymer for various subphase pH values (π = 5 mN/m).
The cross-linked material could be here again removed from the interface with a spatula
after its compression to a final area of ca. 2 x 15 cm2 (Figure 5-20), resulting in a film
approximately 50 monolayers thick. This material had the same physical appearance as the cross-
linked hydrosilylated PB homopolymer, it was insoluble in common organic solvents, self-
119
supporting and gel-like unlike the non-hydrosilylated (PB-b-PEO)3 star block copolymer, and
could also be collected as elongated elastic sheets drawn into very long fibers at high elongation.
Figure 5-20. Removal of the cross-linked (PB(Si(OEt)3)-b-PEO)3 three-arm star copolymer from
the Langmuir trough surface.
5.2.2.3 AFM imaging
The morphologies of the LB films transferred onto mica substrates after cross-linking at
different surface pressures were characterized by AFM. (Figure 5-21). All the transfer ratios
120
were close to unity, so it is assumed that the morphologies observed in Figure 5-21 are not
modified during LB film formation. In a control experiment for a LB film prepared prior to any
significant cross-linking takes place (π = 5 mN/m, pH = 7.0), a smooth and featureless
monolayer with no phase separation between the hydrosilylated PB blocks and the PEO blocks
was obtained (Figure 5-21A). After the hydrosilylation reaction, the PB blocks become more
hydrophilic because of the triethoxysilane pendant groups and are therefore adsorbed at the A/W
interface like the PEO blocks. This interfacial property of the hydrosilylated PB block was
already demonstrated in the preliminary investigation on the hydrosilylated PB homopolymer
which formed stable monolayers for surface pressures as high as 40 mN/m. Such a behavior
differs significantly from the PB block of the (PB-b-PEO)3 star block copolymer which is much
more hydrophobic and aggregates above the water surface. When the hydrosilylated star block
copolymer was reacted under isobaric conditions for 10 h at 2 mN/m (Figure 5-21B), a clear
phase separation between the cross-linked PB material (yellow areas) and the PEO chains (dark
areas) was observed, with an average height of about 2 nm for the cross-linked domains.
However, it is only for surface pressures equal or higher than 6 mN/m that true PEO pores are
trapped within the PB network (Figure 5-21D). As the surface pressure is further increased, the
average PEO pore size decreases (Figures 5-21D, 5-21E, 5-21F, and 5-21G) to reach a
morphology with very small PEO pores (π = 9 mN/m; Figure 5-21G). For even higher surface
pressures such as 15 mN/m (Figure 5-21H), the cross-linked PB covers the entire surface with
the PEO pores barely visible. This is in good agreement with the fact that, at 10 mN/m, the PEO
chains are pushed inside the water subphase. This was confirmed by cross-section analysis which
showed that the monolayer cross-linked at 15 mN/m, with the PEO chains dissolved in the
aqueous subphase underneath the PB network, is much smoother (smaller signal amplitude,
121
Figure 5-22B) than the one cross-linked at 6 mN/m (Figure 5-22A) with the PEO segments still
present at the interface. The average sizes of the PEO pores were roughly determined from the
power spectral densities (PSD) (Figures 5-22C and 5-22D) of the AFM images.225 The
wavelengths corresponding to the peaks in the PSD plots give the average distance between
nearest neighbor PB domains for the monolayer transferred at 5 mN/m (≈ 180 nm) and the
average PEO pore size for the monolayers transferred at 6, 9, and 15 mN/m. This characteristic
pore size decreases from 130 nm for π = 6 mN/m down to 40 nm for π = 9 mN/m (Figure 5-
22C). The PSD curves of the images of the LB films transferred at 7 and 8 mN/m were not
included for easier visualization, but the average PEO “pore” sizes were equal to 46 and 42 nm,
respectively. For surface pressures higher than 10 mN/m, no maxima were observed in the PSD
plots, which is in good agreement with the fact that the PEO pores could hardly be seen for such
surface pressures as shown in Figure 5-21H. It can be concluded from this AFM analysis that the
cross-linking reaction takes place homogeneously on the water surface, allowing the formation
of a 2D network with PEO pores of controllable sizes by simply adjusting the polymerization
surface pressure.
Another experiment to illustrate the possibility to retain a specific morphology after cross-
linking was attempted. A (PB(Si(OEt)3)-b-PEO)3 monolayer was cross-linked at 9 mN/m for 10
h (pH = 3.0) and transferred onto mica. A second transfer of the same cross-linked material was
performed after barrier expansion back to 2 mN/m. As shown in Figure 5-23, the two LB films
have similar morphologies, with only a slight increase in PEO pore size after monolayer
expansion.
122
Figure 5-21. AFM topographic images of the (PB(Si(OEt)3)-b-PEO)3 star block copolymer LB films. (A): t = 0 h. (B), (C), (D), (E), (F), (G), and (H): t = 10 h. The images are 2 x 2 µm2.
Figure 5-22. (A) and (B): Cross-section analysis of Figures 5-21D and 5-21H. (C): PEO pore size versus surface pressure plot. (D): PSD plots of Figures 5-21C, 5-21D, 5-21G, and 5-21H.
123
Figure 5-23. AFM topographic images and corresponding cross-sections of the (PB(Si(OEt)3)-b-
PEO)3 star block copolymer LB films cross-linked at 9 mN/m (pH = 3.0, t = 10 h) and transferred at 9 and 2 mN/m. The images are 2 x 2 µm2.
A final experiment was designed to prove that, when the cross-linking reaction is carried
out above 10 mN/m, the PEO chains are irreversibly dissolved and held into the aqueous
subphase, underneath the cross-linked PB network. After cross-linking the monolayer at 20
mN/m (t = 10 h, pH = 3.0), the barriers were fully expanded and the isotherm of the cross-linked
monolayer was recorded as shown in Figure 5-24 (blue curve). The PEO-related phase transition
(pseudoplateau) is no longer present, which confirmed that the PEO chains could not readsorb at
the interface during monolayer expansion. A control experiment was carried out by recording the
isotherm of a monolayer cross-linked below the surface pressure corresponding to the PEO
aqueous dissolution (5 mN/m, Figure 5-24, red curve). As expected, the PEO pseudoplateau is
still present (even after several compression-expansion hysteresis cycles), which confirms that it
is possible at high surface pressure (π > 10 mN/m) to freeze the “bilayer” conformation of the
cross-linked material consisting of a cross-linked PB layer covalently attached to a PEO
sublayer.
124
Figure 5-24. Surface pressure-MMA isotherms of the hydrosilylated (PB(Si(OEt)3)-b-PEO)3 star
block copolymer cross-linked at 5 and 20 mN/m (pH = 3.0, t = 10 h).
5.3 Conclusions
The main objective of this study was to propose a new and general method to synthesize a
novel 2D polymeric nanomaterial consisting of a continuous cross-linked PB network containing
PEO pores of controllable sizes. To reach that goal, a novel (PB(Si(OEt)3)-b-PEO)3 three-arm
star block copolymer was synthesized by hydrosilylating the PB pendant double bonds of a (PB-
b-PEO)3 three-arm star block copolymer with triethoxysilane. Spontaneous hydrolysis and
condensation under acidic conditions of the triethoxysilane pendant groups of the (PB(Si(OEt)3)-
125
b-PEO)3 three-arm star block copolymer allowed the easy cross-linking of the PB segments
directly at the A/W interface without any additives or reagents. This strategy permits to control
the size of the PEO pores simply by adjusting the surface pressure during cross-linking as shown
by AFM imaging of the LB films.
Further characterization of these 2D cross-linked networks will be required (gas
permeability, small angle scattering, and 2D viscometry) to understand the benefits provided by
the A/W interfacial self-assembly compared to other conventional solution self-adsorption or
other processes. At stake is the possibility to use 2D self-organization as a means to construct
materials with anisotropic structures, to reproducibly engineer such structures, and to target
defined functions with these materials. In addition, such alkoxysilane-containing monolayers
could also be easily grafted onto inorganic surfaces (glass support such as silicon wafer) through
covalent bonds to synthesize polymer/inorganic composite materials.
5.4 Experimental Methods
5.4.1 Materials and Instrumentation
The synthesis of the PB-b-PEO three-arm star block copolymers was previously
reported.222 Toluene used in the hydrosilylation reactions was dried and distilled twice over CaH2
and polystyryllithium successively. The PB homopolymer (Mn = 11,050 g/mol; Mw/Mn = 1.04)
(Polymer Source Inc.), triethoxysilane (Aldrich, 99%), and platinum(0)-1,3-divinyl-1,1,3,3-
tetramethyldisiloxane complex (Karstedt catalyst, 3 wt % solution in xylene) (Aldrich, 99%)
were used as received without further purification. 1H-NMR spectra were recorded on Varian-
VXR 300 MHz, Gemini 300 MHz, and Mercury 300 MHz using CDCl3 as the deuterated
solvent. Chemical shifts are reported in ppm (δ) downfield from tetramethylsilane (TMS) and
referenced to residual chloroform (7.27 ppm). FTIR absorbance spectra were recorded on a
Brüker/Vector 22 FT/IR spectrometer. The samples were prepared by dissolving the
126
hydrosilylated and non-hydrosilylated polymers in chloroform (C ≈ 1 mg/mL), and the resulting
solutions were subsequently spread dropwise onto KBr pellets and allowed to dry under vacuum
in a desiccator.
5.4.2 Langmuir Films
Surface film characterization was accomplished by use of a Teflon Langmuir trough (W =
150 mm and L = 679 mm) system (KSV Ltd., Finland) equipped with two moving barriers and a
Wilhelmy plate for measuring surface pressure. Between runs, the trough was cleaned with
ethanol and rinsed several times with Millipore filtered water of ∼ 18 MΩ.cm resistivity. The
subphase temperature was maintained at 25 °C through water circulating under the trough.
Samples were typically prepared by dissolving ∼1 mg of polymer in 1 mL of chloroform and
spread dropwise with a gastight Hamilton syringe on the Millipore water surface. For the surface
property studies of the non-hydrosilylated (PB-b-PEO)3 star block copolymers, the chloroform
was allowed to evaporate for 30 min to ensure no residual solvent remained, and the isotherms
were subsequently recorded with barrier movement of 5 mm.min-1. For the surface property
studies of the (PB(Si(OEt)3-b-PEO)3 three-arm star block copolymer and the triethoxysilane-
functionalized PB homopolymer, the isotherms were recorded after different reaction times at
zero pressure (pH = 3.0, barrier compression speed of 100 mm.min-1). The isobaric experiments
were carried out just after spreading and after monolayer compression up to a surface pressure of
5 ((PB(Si(OEt)3-b-PEO)3 three-arm star block copolymer) or 10 mN/m (triethoxysilane-
functionalized PB homopolymer) for different subphase pH values.
5.4.3 Hydrosilylation of the PB Homopolymer
In a flame and vacuum-dried three-neck flask equipped with a magnetic stirrer, a reflux
condenser, a dry argon inlet, and a heating mantle, 142 mg (1.28x10-5 mol) of linear PB (Mn =
127
11,050 g/mol) were freeze-dried and then dissolved in 20 mL of dry toluene. Triethoxysilane
(0.573 mL, 3.133x10-3 mol) and Karstedt catalyst (0.1 mL, 3 wt % solution in xylene) were then
added, and the reaction was carried out under argon for 24 h at 80 °C. At the end of the reaction,
the solvent and the unreacted triethoxysilane were removed by evaporation under vacuum, and
the hydrosilylated PB homopolymer was stored under dry argon (Crude product: Mn (1H NMR)
= 33,200 g/mol; m = 135 mg). 1H NMR (δppm; CDCl3): 5.4 (bm, —CH2—CH=CH—CH2— and
CH2=CH—CH—), 4.9 (b, CH2=CH—CH—), 3.8 (b, —CH2—Si(OCH2CH3)3), 2.0 (bm, —
CH2—CH=CH—CH2— and CH2=CH—CH—), 1.5-0.7 (b, —CH2—Si(OCH2CH3)3, and
CH2=CH—C(R)H—CH2—), and 0.3-0.7 (b, —CH2—Si(OEt)3).
5.4.4 Hydrosilylation of the (PB200-b-PEO326)3 Three-Arm Star Block Copolymer
In a flame and vacuum-dried three-neck flask equipped with a magnetic stirrer, a reflux
condenser, a dry argon inlet, and a heating mantle, 164 mg (2.17x10-6 mol) of (PB200-b-PEO326)3
three-arm star block copolymer were freeze-dried and dissolved in 15 mL of dry toluene.
Triethoxysilane (0.285 mL, 1.562x10-3 mol) and Karstedt catalyst (platinum(0)-1,3-divinyl-
1,1,3,3-tetramethyldisiloxane complex, 0.1 mL, 3 wt % solution in xylene) were added, and the
reaction was carried out under argon for 24 h at 80 °C. At the end of the reaction, the solvent and
the unreacted triethoxysilane were removed by evaporation under vacuum, and the product was
stored under argon (Crude product: Mn (1H NMR) = 135,500 g/mol; m = 160 mg). 1H NMR
(δppm; CDCl3): 5.4 (bm, CH2—CH=CH—CH2— and CH2=CH—CH—), 4.9 (b, CH2=CH—
CH—), 3.8 (b, —CH2—Si(OCH2CH3)3), 3.6 (b, —CH2—CH2—O—), 2.0 (b, CH2—CH=CH—
CH2— and CH2=CH—CH—), 1.5-0.9 (b, —CH2—Si(OCH2CH3)3, and CH2=CH—C(R)H—
CH2—), and 0.3-0.8 (b, —CH2—Si(OEt)3).
128
5.4.5 A/W Interfacial Cross-Linking
Chloroform solutions (100 µL, C = 1 mg/mL) of the (PB(Si(OEt)3)-b-PEO)3 three-arm star
block copolymer or of the hydrosilylated PB homopolymer were spread dropwise with a gastight
Hamilton syringe on the Millipore water subphase at pH = 3.0. The monolayers were
immediately compressed with barrier compression speed of 100 mm.min-1 and held at the desired
surface pressure. In the case of the (PB(Si(OEt)3)-b-PEO)3 three-arm star block copolymer, the
cross-linking reaction was carried out for 10 h, and the cross-linked monolayers were
subsequently transferred onto mica substrates (transfer rate = 1 mm/min) for further AFM
characterization.. In the case of the hydrosilylated PB homopolymer, the Langmuir films were
transferred onto freshly cleaved mica after cross-linking at 10 mN/m for various reaction times.
The LB films were dried in a dessicator for 24 h and subsequently scanned in tapping mode with
a Nanoscope III AFM (Digital Instruments, Inc. Santa Barbara, CA) using Nanosensors silicon
probes (dimensions: T = 3.8-4.5 µm, W = 27.6-29.2 µm, L = 131 µm). The images were
processed with a second order flattening routine (Digital Instruments software). After cross-
linking at 15 mN/m for at least 10 h to ensure completion of the cross-linking reaction, further
compression of the resulting cross-linked monolayers (cross-linked (PB(Si(OEt)3)-b-PEO)3
three-arm star block copolymer or hydrosilylated PB homopolymer ) to a final area of ca. 2 x 15
cm2 led to the formation of multilayers (≈ 50 monolayers) that were easily removed with a
spatula from the water surface and dried in a dessicator for further FTIR study and solubility
experiments.
129
CHAPTER 6 ELECTROCHEMICAL AND SPECTROSCOPIC CHARACTERIZATION OF ORGANIC
COMPOUND UPTAKE IN SILICA CORE-SHELL NANOCAPSULES
6.1 Introduction
We recently prepared a series of nanocapsules made of core-shell silica spheres filled with
a hydrophobic solvent, such as ethyl butyrate.49,226 Such nanocapsules are interesting objects in
that the core is fluid and therefore highly dynamic, while the shell is solid and mesoporous, and
appropriate surface modification leads to aqueous dispersions that are stable for months. In
addition, the nanocapsules have sizes that can be controlled from some 60 to 600 nm, have
extremely high interfacial area, have moderate polydispersity, and efficiently absorb
hydrophobic compounds including drugs. These and other submicron-sized particles have shown
a great potential for encapsulation of guest molecules. Representative examples of other systems
include solid lipid nanoparticles (SLNs),227,228 nanoparticles,229-231 microspheres,232,233
liposomes,234,235 polymer vesicles,236 host-guest carriers,237,238 and shell cross-linked Knedel-like
nanoparticles (SCKs).239,240
As introduced in Chapter 1, the current investigations are aimed toward harnessing the
encapsulation abilities of the nanocapsules for potential drug detoxification applications.
Analogous controlled drug release applications often require the release of drug from the particle
to be slow, zero order, and stable with time.232 Furthermore, the relative permeability of different
species within the particle is not often considered. Drug detoxification, however, requires that
the uptake of drug into the particle be as rapid as possible, and often further uptake beyond tens
of minutes would be less desirable. Furthermore, the selectivity of the particle to the drug should
be as high as possible, and therefore the permeability of the particle to substances of different
molecular sizes and polarizabilities should be minimal.
130
In this chapter, we demonstrate that combined electrochemical (cyclic voltammetry) and
optical (UV-vis and fluorescence spectroscopies) techniques provide a useful way to analyze the
uptake of lipophilic substances by the nanocapsules and like systems. The primary goal of this
study is to fundamentally understand the host-guest interactions and the physical properties
driving the encapsulation process using simple analytical techniques. In this work, we therefore
decided to use common organic probes suitable for optical and electrochemical experiments
(iodine, Nile Red, ferrocene methanol, and ferrocene dimethanol), because while UV-vis
spectroscopy can be applied to a variety of toxic drugs,241,242 fluorescence spectroscopy and
electrochemistry are more restrictive techniques. The results discussed in this chapter will
subsequently be of great importance in designing core-shell nanoparticulate systems for efficient
and selective encapsulation applications.
The essence of the electrochemical measurement is simply that any electroactive substance
absorbed into a nanocapsule shows no or very little electrochemical activity, which leads to a
decrease in the measured electrochemical signal upon encapsulation. The origin of this negligible
electrochemical activity of the encapsulated probes compared to the ones remaining in solution
can be rationalized by the following two theories. Upon encapsulation, the electrochemical probe
might stop behaving as an electroactive material because the core and shell of the particle disrupt
electron transfer processes. In fact, previous work has shown that a compact monolayer of the
size of only 12-14 CH2 groups was sufficient to block the electron transfer between an
electroactive substance and an electrode,243,244 while the shell thicknesses of our nanocapsules are
much larger than this. Extensive work has also been reported on dendrimers where the electron
transfer was attenuated as the number of generations increases around redox-active cores.245,246
The other theory that would rationalize the negligible electrochemical activity of the
131
encapsulated probes relies on the fact that their diffusion coefficient toward the working
electrode is essentially similar to the diffusion coefficient of the encapsulating agent as it was
extensively shown for micellar systems.247-250 With encapsulating agents as big as the
nanocapsules studied here, the apparent diffusion coefficient of the encapsulated probes is
sufficiently decreased, so their electrochemical signal can be neglected as quantitatively shown
under Results and Discussion. Both theories lead to the same experimental observation: a
negligible influence of the encapsulated probes on the overall electrochemical signal.
Nevertheless, additional experiments will be necessary to determine which one of these two
theories best describes our encapsulating nanoparticulate system.
All experiments were performed following the same strategy. An aqueous solution
containing a well-defined initial concentration of an electroactive or fluorescent probe was added
to a given amount of nanocapsules. Since the electrochemical signal of the encapsulated probe
molecules is negligible, the uptake can be monitored relative to the elapsed time by the
difference between the initial electrochemical signal and that after uptake. When spectroscopy
(fluorescence or UV-vis) was used instead of electrochemistry, the post-uptake location of the
probe within the nanocapsules could be determined by monitoring the signal maximum
wavelength, which is strongly dependent on the probe chemical environment.
6.2 Results and Discussion
The nanocapsule synthesis is summarized in Figure 6-1. The goal of this work was to
evaluate the efficiency of the uptake of various hydrophobic molecules and to analyze how the
shell thickness influences the uptake kinetics. The samples prepared for the work thus had
similar compositions in terms of octadecyltrimethoxysilane (OTMS), lecithin, Tween-80, and
ethyl butyrate amounts, but the tetramethoxysilane (TMOS) weight percent used was increased
from 0.07 up to 0.88 wt %, allowing the synthesis of nanocapsule samples with different shell
132
thicknesses (Figure 6-2). The core and the shell sizes were characterized as detailed in the
following, both by dynamic light scattering (DLS) and transmission electron microscopy (TEM)
after a preliminary staining step so that the core and the shell of the nanocapsules could be
distinguishable.
Figure 6-1. Oil-filled silica nanocapsule synthesis through initial hydrophobic core formation
followed by hydrophilic silica shell formation after TMOS addition.
Figure 6-2. Description of the nanocapsule samples prepared using 0.07 wt % TMOS (a), 0.28 wt
% TMOS (b), 0.44 wt % TMOS (c), and 0.88 wt % TMOS (d).
133
6.2.1 Nanocapsule Characterization
Figure 6-3 shows as an example the DLS results of the thin shell nanocapsule samples.
Curve a, which represents nanocapsules prior to TMOS addition, shows two peaks. The one
around 7-8 nm corresponds to the diameter of micelles arising from Tween-80 aggregation, and
the other one around 40 nm corresponds to the oil core diameter of the nanocapsules after
microemulsion formation. Curve b corresponds to 0.07 wt % TMOS added to the microemulsion
analyzed in curve a. The peak around 40 nm shifted to approximately 80 nm, showing shell
formation (≈ 20 nm thick) upon TMOS addition. The peak around 7-8 nm disappeared because
the samples have been dialyzed before the DLS measurement is run, showing that almost all the
Tween-80 micelles have been removed. This is a particularly important point for the following
study, because in the case micelles were present they could also be responsible for one part of the
active molecules uptake, and therefore no sensible conclusion could be drawn from it. In the rest
of the study, any possible micelle influence has therefore been purposely avoided.
Overall, the particle size analysis shows relatively narrow distributions and, as expected, a
proportional relationship between the shell size and the amount of TMOS used, leading to
nanocapsules diameters in the range of 80-200 nm, according to the synthetic conditions.
Figure 6-3. DLS results for the microemulsion immediately after preparation (a) and the same solution after TMOS addition (0.07 wt %) and dialysis (b).
134
The nanocapsules were also characterized by TEM analysis. Figure 6-4 shows examples of
particles with 0.07 wt % TMOS (a) and 0.88 wt % TMOS (b). The nanocapsule core has been
stained with 1-dodecene and osmium tetroxide (OsO4). As a consequence, only the hydrophobic
core absorbing 1-dodecene is stained and appears darker than the silica shell. We should note
that, due to the high solubility of Tween-80 in ethyl butyrate, some core stain also comes from
the reaction between Tween-80 and OsO4. As shown in Figure 6-4, the nanocapsule
characteristic sizes (core diameter/shell thickness) are in good agreement with the DLS results.
Figure 6-4. TEM micrographs of the 0.07 wt % TMOS nanocapsules (a) and of the 0.88 wt % TMOS nanocapsules (b).
6.2.2 Uptake Study
Two types of electroactive molecules have been used for the study. Their choice was based
on the fact that such probe molecules must show perfectly reversible redox behavior over many
cycles, must have a strong affinity for the organic phase inside the nanocapsules, and yet must
retain aqueous solubility sufficient for the initial electrochemical measurements to be possible.
Therefore, we decided to use ferrocene methanol and ferrocene dimethanol (Figure 6-5), since
these simple probes are soluble in both organic solvents and water. Due to the presence of the
135
hydroxyl groups, both ferrocenes may show a tendency to bind reversibly through hydrogen
bonding to the nanocapsule silica shell, and not only to partition in the oil phase, but also to
reside partly in the silica shell. It has been previously shown that the diffusion coefficients of
these two molecules in hydrated Sol-Gel-derived glasses were only slightly decreased from the
ones in solution with values in the range of 10-6 cm2/s.251 Therefore, in the presence of diffusion-
driven encapsulation, even with the presence of specific interactions with the silica shell, fast
uptake kinetics would be expected to be seen, because the average time for a molecule to diffuse
through a few tens of nanometers (using the simple relationship252 2/1)2( Dtd = , with t = time of
the experiment, D = diffusion coefficient of the probe, and d = average distance travelled by the
probe during t) is on the order of 10-6 s.
Absorption and fluorescence spectroscopic studies were also undertaken in order to
supplement and confirm the electrochemical study. The two molecules chosen in this case were
iodine and Nile Red (Figure 6-5). Because of their structures, they should have very low affinity
for the silica shell and are therefore expected to partition strictly between the oil core and the
aqueous solution. Because UV-vis and fluorescence measurements show a chemical-
environment-dependent signal maximum, the shift observed after nanocapsule solution addition
to the iodine or Nile Red solution indicates the environment where the probes end up after
encapsulation.
Figure 6-5. Chemical structures of ferrocene methanol, ferrocene dimethanol, and Nile Red.
136
6.2.2.1 Optical measurements results
Figure 6-6 represents the UV-vis absorption spectra of iodine in various aqueous and
organic environments. The measurements were recorded after addition of 1mL of a saturated
iodine aqueous solution to given amounts of either nanocapsule or ethyl butyrate solutions. The
large signal shift toward lower wavelengths (blue shift) compared to the spectrum in water
qualitatively indicates that the chemical environment of iodine significantly changes (becomes
more hydrophobic) after addition of the nanocapsule solution, and that therefore a consequent
uptake in the hydrophobic nanocapsule core took place. However, test experiments have also
been performed on a Tween-80 aqueous solution to appreciate the influence of this compound in
the process. The overlap of the iodine UV-vis curves in the presence of nanocapsules and
Tween-80 indicates that Tween-80 alone was also capable of absorbing/encapsulating iodine in
its hydrophobic micellar core (C18 chains), so that it is not very clear at this point if a separate
layer of Tween-80 in the nanocapsules, Tween-80 dissolved in the organic solvent core, or the
organic solvent core itself is effective in retaining iodine.
Figure 6-6. UV-vis absorption spectra of iodine in water solution (a), in nanocapsule solution (b), in Tween-80 aqueous solution (c), and in ethyl butyrate solution (d).
137
In another experiment, Nile Red was used as a fluorescent probe, because fluorescence is a
more sensitive technique than UV-vis absorption. This classical dye is fluorescent and soluble in
aqueous medium only at acidic pH values. Several fluorescence experiments have been carried
out in order to check the uptake and the environment of the dye after nanocapsule absorption.
Fluorescence spectra of Nile Red were measured in acidic water solution, in nanocapsule
solutions, in a Tween-80 aqueous solution, in ethyl butyrate solution, and on silica gel under
various conditions as shown in Figure 6-7.
Figure 6-7. Nile Red emission spectra in ethyl butyrate solution (a), in nanocapsule solution (b), in Tween-80 aqueous solution (c), in crushed Xerogel dispersion in acidic water (d), on silica gel (e), and in acidic water solution (f).
In this set of experiments, the dye can be selective for comparison between the two types of
silica environments, the acidic water, the Tween-80, and the ethyl butyrate. The data show that
both the silica and the acidic water give the same fluorescence signal, which corresponds to the
acidic form of Nile red. On the other hand, in the presence of Tween-80 or the nanocapsules, a
fluorescence signal typical of Nile Red in an organic environment is obtained, which proved the
complete uptake of the dye by both the Tween-80 micelles and the nanocapsules. It should be
138
noted that Nile Red in the nanocapsules is probably in a Tween-80 environment (on the inner
shell wall) and not in the ethyl butyrate core, because the spectrum is almost the same as in
Tween-80 micelles and different from that in the ethyl butyrate solution.
6.2.2.2 Electrochemical experiments
Ferrocene methanol and ferrocene dimethanol uptakes were measured as a function of time
by electrochemistry. As a preliminary experiment, we determined the partition coefficients of
these two compounds between ethyl butyrate and water by UV-vis spectroscopy (by stirring
biphasic solutions with a given amount of ferrocene methanol or dimethanol and measuring the
absorbance in the aqueous phase after stabilization). As expected, ferrocene methanol and
ferrocene dimethanol turned out to be more soluble in ethyl butyrate than in water, with partition
coefficient values of 40 and 2.5, respectively.
The experiments involved the use of cyclic voltammetry to evaluate changes in the
concentration of free electroactive species. An example of a typical cyclic voltammogram
recorded for ferrocene methanol is shown in Figure 6-8.
Figure 6-8. Typical cyclic voltammogram of ferrocene methanol in water.
139
The peak potentials difference is around 60 mV. As discussed in Chapter 2, this shows
that, at this potential scan rate (500 mV/s), ferrocene methanol and ferrocene dimethanol
conversions remain reversible. The currents therefore remain basically diffusion controlled, and
the faradaic current (Ip, difference in intensity between the oxidation peak and the residual
current) is given by the following Randles-Sevcik equation (6-1) previously introduced in
Chapter 2,
2/12/12/35 )10*69.2( vACDnI p = (6-1)
where C is the concentration of the electroactive species, A is the surface of the working
electrode in cm2, D is the diffusion coefficient of the electroactive species in cm2/s, ν is the
potential scan rate in V/s, and n is the number of electrons transferred in the redox process. A, n,
and ν are kept constant during the cyclic voltammetry scans, and therefore Ip is directly
proportional to D1/2C. As mentioned in the introduction of this chapter, when the nanocapsules
are added, the encapsulated molecules become either electrochemically inactive (electron
transfer cannot occur within distances larger than 5-10 nm and the particle shell is always larger)
or have an apparent diffusion coefficient sufficiently low so that the signal intensity arising from
the encapsulated probes can be neglected. The diffusion coefficient D for spherical particles is
given by the following Stokes-Einstein equation (6-2) previously introduced in Chapter 2,
RTkD B
πη6= (6-2)
where kB is the Boltzmann constant (kB =1.38 × 10-23 m2 kg s-2 K-1), T is the temperature in K, η
is the solution viscosity in P, and R is the particle radius in m. When this equation is applied to
our nanocapsules of 40-100 nm radius, the diffusion coefficient is calculated in the range (2-
6)x10-8 cm2/s, which is a factor 102-103 smaller than a typical diffusion coefficient for a single
molecule like ferrocene methanol or dimethanol (in the range 10-5-10-6 cm2/s). The signal from
140
the nanocapsules is therefore expected to be beyond detections, even in the unlikely case of a
non-negligible electron transfer within the shell. It should be noted here that the reversibility of
the uptake in principle could be observable and studied, upon conversion of most of the
ferrocene into ferricinium ion inside the water. However, this requires extensive electrolysis of
the remaining ferrocene inside the water, which is complicated, not only because it implies using
a cell allowing both analytical electrochemistry and extensive electrolysis, but above all because
the stability of the ferricinium ion is not sufficient, especially in water (it slowly undergoes
nucleophilic attack from water).
The electrochemical signal therefore quantitatively decreases in proportion to the uptake,
and the aqueous concentration can be calculated by applying the relationship Ip α C. Since the
initial concentration in electroactive compounds in the solution is accurately known, it is not
necessary to know the exact electrode parameters that would make the measurements less
precise. The relative variation of the concentration, and therefore the electroactive compound
uptake inside the nanocapsules, is directly obtained. The partition coefficients between Tween-
80 and water for ferrocene methanol and ferrocene dimethanol could not be measured because of
the high viscosity of Tween-80 and its high solubility in water. Nonetheless, we suppose that,
analogous to the behavior observed in the spectroscopy experiments above, the Tween-80
present in the oil core might increase the uptake.
The ferrocene methanol uptakes for the nanocapsule samples introduced earlier are shown
in Figure 6-9. The normalized aqueous concentration decrease is plotted versus time. All the
nanocapsule samples show significant decrease in the ferrocene methanol concentration, and as
might have been expected, the overall uptake is roughly increased with increasing nanocapsule
concentration (Figure 6-10). Contrary to the iodine and the Nile Red spectroscopic experiments
141
where the signal shift was almost instantaneous upon addition of the nanocapsule solutions, we
see slower uptake kinetics for the larger shells, indicating a shell-size influence. Ferrocene
methanol interacts with the silica shell through hydrogen bonding interactions in a phenomenon
similar to that observed in classical column chromatography via an adsorption/desorption
mechanism. As previously reported, such a mechanism reduces the diffusion coefficient in the
silica shell, but in order to observe simply a slower diffusion-driven uptake, the diffusion
coefficient in the nanocapsule shell should in theory be reduced by several additional orders of
magnitude (in the 10-12-10-13 cm2/s range). Such a large decrease of the diffusion coefficient in
the shell is unlikely for a classical diffusion mechanism; therefore, other additional shell-
thickness-dependent parameters must be involved.
Figure 6-9. Uptake of ferrocene methanol versus time in 0.07 wt % TMOS nanocapsule solution (b1, total concentration: 5.9 wt %; b2, total concentration: 6.2 wt %), in 0.28 wt % TMOS nanocapsule solution (c, total concentration: 1.9 wt %), in 0.44 wt % TMOS nanocapsule solution (d, total concentration: 1.9 wt %), and in 0.88 wt % TMOS nanocapsule solution (e, total concentration: 4.9 wt %). (a) Control experiment in the absence of nanocapsules. The dotted lines are provided to highlight the trends.
142
Figure 6-10. Plot of normalized aqueous concentration of ferrocene methanol after uptake in 0.07 wt % TMOS nanocapsule solution (b1, total concentration: 5.9 wt %; b2, total concentration: 6.2 wt %), in 0.28 wt % TMOS nanocapsule solution (c, total concentration: 1.9 wt %), in 0.44 wt % TMOS nanocapsule solution (d, total concentration: 1.9 wt %), and in 0.88 wt % TMOS nanocapsule solution (e, total concentration: 4.9 wt %). (a) Control experiment in the absence of nanocapsules.
While further quantitative analysis is somewhat difficult to make, support from this
interpretation comes from the fact that the only cases where appreciable slower kinetics are seen
in Figure 6-9 are the two cases where the largest amount of TMOS was introduced in the
synthesis solution (curves d and e that respectively correspond to nanocapsule samples with 0.44
and 0.88 wt % TMOS), and therefore where thicker shells were obtained. For these two samples,
it takes approximately 200-300 s before the aqueous concentration of ferrocene methanol levels
off, whereas the thermodynamic equilibrium is reached almost instantaneously for the other three
samples with thinner silica shells (curves c and b that respectively correspond to nanocapsule
samples with 0.28 and 0.07 wt % TMOS).
To further investigate silica shell/alcohol interactions, we carried out similar experiments
with ferrocene dimethanol. This substance displays a similar electroactivity with a slight shift in
143
potential (which does not influence this type of measurement) compared to ferrocene methanol,
but contains two alcohol groups, making it not only more hydrophilic but above all much more
likely to interact with the silica shell. The uptake results are presented in Figure 6-11. As
expected from the results with ferrocene methanol, much slower kinetics were observed for all
the nanocapsule samples, including those with thin silica shells. Moreover, compared to the
results for ferrocene methanol, the overall uptake is lower, this due to a higher hydrophilicity of
ferrocene dimethanol. However, one must notice that this is not true for the 0.28 wt % TMOS
nanocapsule sample, where the uptake is higher for ferrocene dimethanol, although the reason
for this result is not very clear. Moreover, based on the partition coefficients mentioned before,
the uptake seems surprisingly large. Even though the concentration of ferrocene dimethanol is
probably smaller in the nanocapsule core, we suppose the significant uptake observed might be
due to increased interactions between the polar ferrocene dimethanol OH groups and the
hydrophilic silica shell, which would be significantly reduced in the case of ferrocene methanol.
The ferrocene dimethanol uptake is therefore more complicated to quantitatively analyze because
of the combined influence of the oil core and the silica shell. The influence of the silica shell on
the overall uptake was already shown to some extent by the results for ferrocene methanol
(Figure 6-10), where samples d (0.44 wt % TMOS, total concentration: 1.9 wt %)) and e (0.88 wt
% TMOS, total concentration: 4.9 wt %) showed increased encapsulation efficiencies compared
respectively to samples c (0.28 wt % TMOS, total concentration: 1.9 wt %) and b1 (0.07 wt %
TMOS, total concentration: 5.9 wt %). While this study gives some hint of the silica shell
influence on the uptake efficiency and kinetics, complete understanding of the encapsulation
mechanism will require further investigation.
144
Figure 6-11. Uptake of ferrocene dimethanol versus time in 0.07 wt % TMOS nanocapsule solution (b, total concentration: 5.9 wt %), in 0.28 wt % TMOS nanocapsule solution (c, total concentration: 1.9 wt %), in 0.44 wt % TMOS nanocapsule solution (d, total concentration: 1.9 wt %), and in 0.88 wt % TMOS nanocapsule solution (e, total concentration: 4.9 wt %). (a) Control experiment in the absence of nanocapsules. The dotted lines are provided to highlight the trends.
As stated earlier, all the nanocapsule samples were extensively dialyzed in order to
eliminate, as much as possible, potential contaminants besides the nanocapsules, and especially
residual Tween-80 micelles. Uptake measurements were also carried out on Tween-80 aqueous
solutions to check if Tween-80 micelles could influence the uptake. Micellar envelopes have
been shown not to influence the electron transfer between the encapsulated species and the
working electrodes, and therefore only the decrease in the diffusion coefficient of the
encapsulated species influences the electrochemical signal.247-250 As determined by DLS, the
radius of Tween-80 micelles is approximately 4 nm, which leads from the Stokes-Einstein
equation to a diffusion coefficient around 6x10-7 cm2/s. It is a factor 10-102 times smaller than
the diffusion coefficient of single molecules like ferrocene methanol and ferrocene dimethanol;
therefore, encapsulation by Tween-80 micelles can easily be seen with cyclic voltammetry as
145
shown in Figures 6-12 and 6-13. Nevertheless, the diffusion coefficient after encapsulation by
Tween-80 micelles is not sufficiently decreased to completely neglect the influence of the
encapsulated molecules on the overall peak intensity as it was the case for the nanocapsules. The
normalized aqueous concentration of ferrocene methanol and ferrocene dimethanol versus
Tween-80 concentration was plotted assuming that the molecules encapsulated by the micelles
do not contribute at all to the overall electrochemical signal. The real aqueous concentration is
slightly smaller than the one reported, and its accurate determination is possible but would
require knowing both the diffusion coefficients of Tween-80 micelles and of the two probes.
However, the encapsulation kinetics are instantaneous in the case of ferrocene methanol, and
very fast (although slightly discernible) in the case of ferrocene dimethanol. The different uptake
kinetics observed with the nanocapsules, in addition to the fact that all the suspensions were
extensively dialyzed, allows us to state unambiguously that the uptake observed and discussed
before was due only to the nanocapsules.
Figure 6-12. Uptake of ferrocene methanol versus time in 0 wt % Tween-80 aqueous solution (a), in 2 wt % Tween-80 aqueous solution (b), in 4 wt % Tween-80 aqueous solution (c), in 6 wt % Tween-80 aqueous solution (d), and in 8 wt % Tween-80 aqueous solution (e). The dotted lines are provided to highlight the trends.
146
Figure 6-13. Uptake of ferrocene dimethanol versus time in 0 wt % Tween-80 aqueous solution (a), in 2 wt % Tween-80 aqueous solution (b), in 4 wt % Tween-80 aqueous solution (c), in 6 wt % Tween-80 aqueous solution (d), and in 8 wt % Tween-80 aqueous solution (e). The dotted lines are provided to highlight the trends.
6.3 Conclusions
In this chapter, a detailed spectroscopic and electrochemical study of the uptake
mechanism of organic chemicals by core-shell nanocapsules was presented. This method shows
promise as a means to determine the efficiency and the kinetics of the uptake process. From
these experiments, we can conclude that an important factor in the incorporation of organic
chemicals in the core of the nanocapsules is the diffusion through the silica shell, which acts
analogously to a chromatographing layer. However, even in the least favorable case studied, the
incorporation time is short and the nanocapsules are efficient removers of large amounts of
organic compounds present in an aqueous solution. This confirms the expected efficiency of the
nanocapsules to remove toxic substances from body liquids, and therefore their potential in
detoxification media.
147
6.4 Experimental Methods
6.4.1 Nanocapsule Synthesis
Synthesis of the core-shell nanocapsules was carried out according to the previous work
reported by our group.226 OTMS (0.09 g, Gelest Inc.), lecithin (0.05 g, Alfa Aesar), and Tween-
80 (2.8 g, Aldrich Chemical Co.) were used as the surfactants, and ethyl butyrate (0.4 g, Aldrich
Chemical Co.) was used as the hydrophobic oil phase. Microemulsion formation was carried out
by adding those four chemicals in saline solution (9 wt % NaCl aqueous solution, 26 mL) under
heating (70 oC) with vigorous stirring for at least 8 h. OTMS polycondensation was carried out at
pH = 3 using a 0.5 M HCl aqueous solution and by vigorous stirring for 30 min. The shell
thickness was controlled by reacting various amounts of TMOS (0.02 g-0.26 g, Gelest Inc.) with
the unreacted silanol groups of the OTMS molecules present at the microemulsion surface. This
last step was carried out after the pH of the nanocapsule solution was increased to neutral
conditions (pH ≈ 7.4) with 0.5 M NaOH and 0.5 M HEPES-buffered solutions. To remove
unreacted OTMS and TMOS, free Tween-80, and lecithin, extensive dialysis was performed
using Spectra/Por molecular porous membrane tubing with a molecular weight cutoff of 6-8000
Da.
6.4.2 Transmission Electron Microscopy
For the TEM experiments, the nanocapsule core was doped with 1-dodecene (Aldrich
Chemical Co.). After deposition of a droplet on a carbon-coated nickel grid (Electron
Microscopy Sciences) and evaporation of the water in a desiccator, the nanocapsule core was
stained by exposure to OsO4 vapors (Aldrich Chemical Co.) in a closed container for at least 4 h.
All TEM images were obtained using a Hitachi H-7000 instrument at 75 kV.
148
6.4.3 Particle Size Analysis
Size analysis was performed before TMOS addition, and after reaction and dialysis. This
allowed determination of the core diameter and shell thickness for every nanocapsule sample
prepared. All samples were diluted to avoid interparticle aggregation and filtered prior to size
analysis using 0.22 µm pore size filters (Fisher Scientific). DLS was used to follow changes in
particle sizes with a Precision Detectors PDDLS/CoolBatch+90T instrument. The data were
analyzed with the Precision Deconvolve32 Program. Measurements were taken at 20 °C at a 90°
scattering angle using a 632 nm laser source. Final sizes were obtained from the average of at
least five reproducible results.
6.4.4 Spectroscopy Measurements
UV-vis absorption spectra were recorded on a UV-vis Varian CARY 500
spectrophotometer, and excitation and fluorescence emission spectra were measured on a SPEX
Fluorolog-3 (Jobin-Yvon). A right-angle configuration was used. The optical density of the
samples was checked to be less than 0.1 to avoid reabsorption artifacts. Iodine and Nile Red are
both commercially available (Aldrich Chemical Co.). The Nile Red stock solution was prepared
by placing 2 mg of Nile Red in 5 mL of methanol, filtering with a 0.22 µm Millipore filter, and
adding the saturated Nile Red methanol solution to 100 mL of acidic water (pH = 1.2).
UV-vis absorption spectra were recorded after mixing thoroughly 1 mL of a saturated
iodine aqueous solution with reference solutions consisting of 2 mL of water, 2 mL of ethyl
butyrate, 2 mL of water with 1 drop of Tween-80, and 1 mL of water with 1 mL of nanocapsule
solution.
Fluorescence spectra were recorded on Nile Red after excitation at 595 nm. For the
measurements involving the three aqueous solutions (nanocapsules, Tween-80 aqueous solution,
and acidic water), fluorescence spectra were recorded after mixing 0.5 mL of the Nile Red stock
149
solution with 0.5 mL of nanocapsule solution in 2.0 mL of water, 1 drop of Tween-80 in 2.5 mL
of water, and 2.5 mL of acidic water, respectively. For the fluorescence spectrum of Nile Red in
ethyl butyrate, 3 mL of ethyl butyrate was mixed with 0.5 mL of the Nile Red stock solution.
The ethyl butyrate layer was then pipetted into a glass cuvette, and the fluorescence spectrum of
the Nile Red in ethyl butyrate was recorded. For the silica gel experiments, a saturated solution
of Nile Red in ethanol was mixed with 2 g of silica gel and the excess solvent was evaporated to
adsorb Nile Red onto the silica gel. Finally, the Xerogel was prepared by allowing the drying of
a mixture of 12.5 mL of Nile-Red-saturated ethanol, 5.6 mL of tetraethoxysilane (Aldrich
Chemical Co.), and 0.9 mL of water with pH 3.4. The resulting gel was subsequently crushed
and placed in acidic water.
6.4.5 Electrochemistry Experiments
The electrochemical studies were performed using an EG&G PAR 273 potentiostat,
interfaced to a PC computer. Cyclic voltammetry experiments were carried out in a three-
electrode 10 mL electrochemical cell. The working electrode used was a 1 mm diameter disk
vitreous carbon electrode polished on a diamond paste covered rotating disk (Presi) prior to each
experiment. The auxiliary electrode was a platinum wire, and the reference electrode (Ag+/Ag
electrode filled with 0.01 M AgNO3) was checked versus ferrocene as recommended by IUPAC.
In our case, E°(Fc+/Fc) = 0.045 V in acetonitrile with 0.1 M tetraethylammonium perchlorate.
Ferrocene methanol and ferrocene dimethanol were purchased from Aldrich Chemical Co., and
lithium perchlorate was purchased from Fluka (puriss).
The ferrocene methanol and ferrocene dimethanol aqueous solutions used contained 2x10-4
mol/L of either electroactive molecule and 10-1 mol/L lithium perchlorate used as the electrolyte.
A typical uptake experiment was carried out as follows: an initial cyclic voltammogram is
recorded for 5 mL of ferrocene methanol or ferrocene dimethanol solution in the electrochemical
150
cell (potential scan rate = 500 mV/s). The encapsulating solution (1 mL of nanocapsule solution
or Tween-80 aqueous solution) is added to the cell, zero time is defined at the point addition is
complete, and the solution is stirred for 5 s with a magnetic stir bar. Further cyclic
voltammograms are then recorded versus time and the faradaic currents are determined for
uptake calculation. The faradaic current at t = 0 s was simply calculated by dividing the faradaic
current obtained in the initial cyclic voltammogram (before addition of the encapsulation
solution is done) by the dilution factor (6/5 for our experiments). Cyclic voltammograms were
not recorded simultaneously with stirring because otherwise the Randles-Sevcik equation used in
calculating the aqueous concentration of the probes during uptake would not be valid. The
experimental data and the error bars plotted correspond respectively to the average value and to
the standard deviation obtained from three different measurements.
151
CHAPTER 7 TOWARD SPECIFIC DRUG DETOXIFICATION AGENTS: MOLECULARLY IMPRINTED
NANOPARTICLES
7.1 Introduction
As discussed in Chapter 1, drug toxicity is a major health concern worldwide.253
Unfortunately, most life-threatening drug intoxications do not have specific antidotes to
overcome the effects of many drugs. Therefore, in the event of illicit drug uses, suicide attempts
or iatrogenic complications, therapy is usually only restricted to stabilizing the patient. To reduce
drug poisoning, it is only recently that a lot of effort has been put into exploring in depth the
unusual properties provided by nanotechnology science with the design of new nanosize objects
that have the ability to reduce the bio-availability of toxic compounds within the body.34,254 Most
toxic drugs are highly hydrophobic and are therefore only slightly soluble in aqueous
environments such as in the blood stream. As a consequence, several nanosystems have been
synthesized with the aim to encapsulate and isolate the toxic drugs through hydrophobic
interactions to significantly decrease their free blood concentration below toxic levels. Some
examples reported in the literature explored for instance the potential of emulsion- and
microemulsion-based systems in sequestering the toxic drugs inside hydrophobic oil
cores.37,49,226,255 Other recent works investigated the possibility of binding the toxic drugs
through π-π interactions inside modified chitosan nanoparticles.256 When designing drug-
encapsulating systems for in-vivo drug detoxification applications, several properties have to be
taken into consideration. As emphasized in Chapter 1, the system should be smaller than the
smallest capillaries (< 5 µm) to move freely in the blood stream, it should have fast (within
seconds or minutes) and high encapsulation capacities, it should be non-toxic (biocompatible),
biodegradable (slowly enough so the aqueous concentration of the drug released in the blood
152
stays below toxic levels), and, above all, it should be specific to the target drug to avoid side
encapsulation of other undesired molecules present in the blood stream. In an attempt to design
nanoparticulate systems with increased specificity and high encapsulation capacities, we report
in this chapter our investigations on the potential for molecularly imprinted nanoparticles to be
used as encapsulating agents in drug detoxification therapy.
The imprinting strategy in this work uses the non-covalent imprinting approach.73,257 The
synthesis of polymers molecularly imprinted in the bulk with various toxic drugs has been
extensively reported.69,258 Moreover, the possibility of synthesizing molecularly imprinted
nanoparticles has been recently demonstrated.259 With a view toward in vivo drug detoxification
applications, we report here for the first time the synthesis of nanoparticles molecularly
imprinted with amitriptyline (Figure 7-1), which is a commonly used tricyclic antidepressant that
may cause cardiac toxicity at high concentration, and the results of their uptake abilities in
aqueous solutions under physiological pH conditions.
Figure 7-1. Chemical structures of amitriptyline and bupivacaine.
7.2 Results and Discussion
Miniemulsion polymerizations have attracted much attention in the past because of their
great potential in synthesizing nanosized spheres with a variety of properties and applications.260
Tovar and co-workers were to our knowledge the first to report about the potential of
miniemulsion polymerization to synthesize molecularly imprinted nanoparticles by the non-
153
covalent technique, where they demonstrated that water soluble binding monomers such as
methacrylic acid (MAA) were quantitatively incorporated inside the nanoparticles, which is a
crucial condition for an efficient imprinting to take place.261 However, the density of the MAA
groups was probably higher in the outer shell of the nanoparticles, as previously described for
other miniemulsion polymerizations involving water soluble monomers.262 This probably led to
an increased binding efficiency, since the imprinted sites were mostly situated near the
nanoparticle surface and therefore more easily accessible for the template molecules during the
rebinding studies.
In our molecularly imprinted nanoparticle synthesis, ethylene glycol dimethacrylate
(EGDMA) was used as the hydrophobic cross-linker, MAA as the binding monomer,
azobisisobutyronitrile (AIBN) as the oil-soluble radical initiator, hexadecane as the highly
hydrophobic agent preventing Ostwald Ripening,83 ethyl butyrate (EB) as the hydrophobic
porogen, and amitriptyline as the template molecule. The relative molar amounts cross-
linker/monomer/template during the imprinting step ranged around 20/4/1 which are commonly
used ratios in molecular imprinting technology.263 It should be noticed that, to demonstrate the
ability of amitriptyline-based molecularly imprinted nanoparticles to be used as detoxification
agents, we deliberately chose to vary only the amounts of cross-linker, monomer, and template
as shown in Table 7-1. Amitriptyline has non-negligible water solubility which makes it able to
freely dissolve in aqueous systems such as in the blood stream. Nevertheless, amitriptyline is
highly hydrophobic with reported partition coefficient values of several thousands between 1-
octanol and water.264 Therefore, under our experimental miniemulsion polymerization
conditions, incorporation of amitriptyline in the miniemulsion oil core during polymerization can
154
be considered quantitative, making the imprinting possible. The imprinting strategy is shown in
Figure 7-2.
Table 7-1. Loading compositions of the miniemulsions.
Sample EGDMA MAA Amitriptyline hydrochloride
MIP1 600 mg (3.03 mmol) N/A N/A
MIP2 550 mg (2.77 mmol) 50 mg (0.581 mmol) N/A
MIP3 500 mg (2.52 mmol) 100 mg (1.162 mmol) N/A
MIP4 600 mg (3.03 mmol) N/A 50 mg (0.159 mmol)
MIP5 550 mg (2.77 mmol) 50 mg (0.581 mmol) 50 mg (0.159 mmol)
MIP6 500 mg (2.52 mmol) 100 mg (1.162 mol) 50 mg (0.159 mmol)
1/ crosslinking
2/ template, porogenand surfactant removal
Template
Binding monomer
Crosslinker Radical initiator
Porogen Stabilizing surfactant
Figure 7-2. The molecular imprinting strategy in miniemulsion polymerization.
The efficiency of the cross-linking reaction was demonstrated by infra-red (FTIR)
spectroscopy. Figure 7-3 shows the FTIR spectra of MIP1 and MIP3, and the FTIR spectrum of
pure EGDMA before reaction is also included for comparison. The absorbances of the peaks at
1640 cm-1 (stretching vibration frequency of the alkenyl C=C double bonds) and at 3100 cm-1
(stretching vibration frequency of the alkenyl C-H single bonds) almost completely vanished for
MIP1 and MIP3, indicating that the doubles bonds of EGDMA and MAA were successfully
155
consumed during the polymerization.265 It is also interesting to notice for MIP3 the appearance of
a broad absorbance peak in the 3400-3600 cm-1 region that corresponds to the stretching
vibration of O-H bonds from carboxylic acid groups, which confirms the successful inclusion of
the MAA monomers in the nanoparticles during miniemulsion polymerization.266
Figure 7-3. IR absorbance spectra of EGDMA, MIP1, and MIP3.
The apparent hydrodynamic diameters of the nanoparticles were determined by dynamic
light scattering (DLS) just after miniemulsion polymerization and after high dilution to avoid
interparticle aggregation. All the samples show monodisperse distributions with average particle
diameters around 220 nm ± 50 nm, independent of the nanocapsule composition. The DLS size
distribution of MIP6 is shown as an example in Figure 7-4. The ability to design particles through
miniemulsion polymerization with diameters in the nanometer range and with therefore large
surface-to-volume ratios is essential for drug detoxification therapy, because it has been
previously shown that increasing the available contact surface by decreasing the size of the
encapsulation entities significantly enhanced the uptake of toxic drugs, probably resulting from a
surface adsorption phenomena. In the present work, the amount of surfactant
156
(dodecyltrimethylammonium bromide, DTAB) stabilizing the miniemulsion droplets during
polymerization was not varied and was kept relatively low to make particles with diameters
around 200 nm. However, miniemulsion technology allows accurate control of nanoparticle size
by varying the amount of stabilizing surfactant used during polymerization.260 With a view
toward increasing the available surface area for drug detoxification applications, synthesis of
nanoparticles with smaller diameters could therefore ultimately be achieved by increasing the
amount of stabilizing surfactant.
Figure 7-4. DLS size distribution of MIP6.
Tapping mode atomic force microscopy (AFM) imaging of the nanoparticles deposited
onto mica substrates confirmed the results obtained from DLS, with no change in particle size
independent of the initial miniemulsion formulation. As an example, typical AFM images of
MIP6 are presented in Figure 7-5. The nanoparticles are again highly monodisperse in size, with
an average measured diameter of 250 nm ± 50 nm, which is slightly larger than the average value
obtained from DLS. As shown in Figures 7-5c (surface plot) and 7-5d (cross-section analysis),
the nanoparticles tend to flatten upon adsorption on the mica substrate, which explains the
slightly overestimated diameter value obtained from AFM characterization compared to DLS. It
157
is also interesting to notice that AFM imaging of the nanoparticles was also possible after
purification, drying, and resuspension by sonication for 10 min in water with no apparent
changes in size and morphology.
Figure 7-5. Tapping mode topographical AFM images (a, b, and c) and cross-section analysis (d) of MIP6.
MIP1, MIP2, and MIP3 were not molecularly imprinted with amitriptyline and were used as
control experiments. Their binding studies for amitriptyline under physiological pH conditions
(HEPES buffered saline solutions, pH ≈ 7.4) are presented in Figure 7-6. All three samples can
very efficiently bind amitriptyline as shown by the very high partition coefficient values
indicated in the captions of Figure 7-6, and the amount of bound amitriptyline increases as the
nanoparticle concentration is increased. More interestingly, the uptake non-negligibly decreases
as the amount of MAA present in the nanoparticles increases. Under physiological pH
conditions, amitriptyline (pKa = 9.4)264b is present in the aqueous phase at approximately 99% in
its protonated form, and the acid groups present on the nanoparticle surface or pore walls are
158
mostly deprotonated. The results in Figure 7-6 clearly show that the uptake is not driven by
electrostatic interactions between the positively charged amine of amitriptyline and the
deprotonated and negatively charged carboxylic acid groups of the nanoparticles. In the case of
non-imprinted nanoparticles, the uptake is driven by non-specific hydrophobic interactions
between the nanoparticles and the hydrophobic aromatic rings and aliphatic carbon chain of
amitriptyline, which is likely adsorbed on the pore walls through its non-polar moiety with the
protonated amine group facing the aqueous solution. The uptake being driven by hydrophobic
interactions instead of hydrogen bondings or electrostatic interactions in water is well known and
has already been described for other molecularly imprinted polymers used for molecular
recognition in aqueous conditions.267 This is not the case when binding studies are carried out in
less polar organic solvents such as dichloromethane268 or toluene,269 since their dielectric
constants are not as high as for water and do not significantly break polar host-guest interactions.
Figure 7-6. Uptake of amitriptyline by the non-molecularly imprinted nanoparticles MIP1 (Kp ∼ 1600), MIP2 (Kp ∼ 1000), and MIP3 (Kp ∼ 1100). The lines are provided to highlight the trends.
Similar binding experiments of amitriptyline were carried out for the molecularly
imprinted nanoparticle samples MIP4, MIP5, and MIP6 (Figure 7-7). MIP6 that does not contain
any polar carboxylic acid groups has a high affinity for amitriptyline, so this confirms the results
159
obtained previously for MIP1, MIP2, and MIP3, where it was observed that the hydrophobic
interactions play a significant role in the binding. Nevertheless, as the amount of carboxylic acids
in the molecularly imprinted nanoparticles is increased, the uptake is significantly increased as
well (partition coefficient values in captions of Figure 7-7), contrary to the non-imprinted
nanoparticles. This clearly indicates that, although polar interactions alone are obviously weaker
than hydrophobic interactions in aqueous solutions, the presence of specific and shape-persistent
amitriptyline recognition sites formed from the weak electrostatic/hydrogen bonding interactions
between MAA and amitriptyline during the imprinting stage significantly increases the uptake,
and consequently the nanocapsules binding specificity. Prior to carrying out the binding studies
and as indicated in the experimental section, the nanoparticles were extensively washed at least
five times with tetrahydrofuran (THF) until no residual amitriptyline could be detected by UV-
vis spectroscopy in the centrifugation supernatants. Nevertheless, it should be noticed that traces
of amitriptyline probably remained after nanoparticle synthesis and template extraction in MIP4,
MIP5, and MIP6 as shown by the slightly lower uptake (and partition coefficient value) of MIP4
compared to MIP1.
Figure 7-7. Uptake of amitriptyline by the nanoparticles molecularly imprinted with
amitriptyline: MIP4 (Kp ∼ 1200), MIP5 (Kp ∼ 1800), and MIP6 (Kp ∼ 2700). The lines are provided to highlight the trends.
160
As a final experiment, binding studies were carried out with bupivacaine (Figure 7-1) on
nanoparticle samples MIP4, MIP5, and MIP6 molecularly imprinted with amitriptyline.
Bupivacaine (pKa = 8.1) is an amide class local anesthetic used in clinical medicine to provide
local or regional anesthesia during surgical procedures, and the necessity in reducing quickly the
free concentration of bupivacaine has also recently attracted much attention.49 The uptake
experiments presented in Figure 7-8 and the partition coefficients provided in the figure captions
confirmed, in the absence of specific imprinting, that the uptake is mainly driven by hydrophobic
interactions, with the nanocapsule samples containing polar carboxylic acid groups binding less
efficiently bupivacaine molecules.
Figure 7-8. Uptake of bupivacaine by the nanoparticles molecularly imprinted with amitriptyline: MIP4 (Kp ∼ 190), MIP5 (Kp ∼ 120), and MIP6 (Kp ∼ 120). The lines are provided to highlight the trends.
7.3 Conclusions
We demonstrated in this chapter that molecularly imprinted nanoparticles prepared by
simple miniemulsion polymerization can efficiently encapsulate large amounts of hydrophobic
toxic drugs such as amitriptyline or bupivacaine. The absence of template molecules during
nanoparticle synthesis prevents the formation of specific binding sites, and as a consequence the
uptake remains mainly driven by non-specific adsorption onto the nanoparticles through
161
hydrophobic interactions. Nevertheless, the combined presence of cross-linkable binding groups
and template molecules during nanoparticle synthesis allows the formation of specific binding
sites that specifically increase the uptake. The ideal system for drug detoxification therapy
requires very high uptake capabilities and 100% specificity. Optimization of a molecularly
imprinted system can be achieved but requires playing with many variables, such as the
molecular imprinting strategy (covalent or non-covalent), the polymerization method, or the
amounts and identities of binding monomers, cross-linkers, and porogens.81 The purpose of this
work was therefore to demonstrate with relatively simple experimental conditions that molecular
imprinting technology has great potential for in-vivo drug detoxification therapy. Molecular
recognition in water with molecularly imprinted polymers started to be explored only very
recently, with the ultimate goal to efficiently mimic host-guest interactions found in natural
aqueous environments. Therefore, further investigations will be necessary to improve the
specificity and decrease the non-specific hydrophobic binding of the molecularly imprinted
nanoparticles in aqueous media under physiological pH conditions. With a view toward drug
detoxification applications, future work should also include investigating the biodegradability of
these water hydrolysable ester-based nanoparticles as well as their biocompatibility after
preliminary surface modification with for instance tethered poly(ethylene oxide) chains.270
7.4 Experimental Section
7.4.1 Materials
Ethylene glycol dimethacrylate and methacrylic acid were purchased from Aldrich
Chemical Co. and were distilled under reduced pressure before use. Dodecyltrimethylammonium
bromide, ethyl butyrate, amitriptyline hydrochloride, bupivacaine hydrochloride, hexadecane,
azobisisobutyronitrile, and HEPES (4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid) were
purchased from Aldrich Chemical Co. and were used as received without further purification.
162
7.4.2 Nanoparticle Synthesis
The nanoparticles were synthesized by traditional miniemulsion polymerization, following
a previously reported method.261 The amounts of DTAB (10 mg, 0.032 mmol), EB (100 mg, 0.86
mmol), hexadecane (30 mg, 0.13 mmol), and AIBN (12 mg, 0.073 mmol) were kept constant,
and only the amounts of EGDMA, MAA, and amitriptyline hydrochloride in the miniemulsion
formulations were varied as previously indicated in Table 7-1. All the chemicals were mixed
with 5 mL of Millipore filtered water in a 10 mL vial, vigorously stirred for 1 h with a magnetic
stir bar, and then sonicated for 5 min. The resulting opaque and milk-like solution was then
transferred into a 5 mL round bottom flask, and the radical polymerization was carried out in an
oil bath for 20 h at 80 oC under vigorous stirring. The nanoparticles were subsequently
centrifuged out of the solution (15 000 rpm, 1 h). They were then redispersed and centrifuged at
least 5 times out of 50 mL of THF to remove the unreacted chemicals and amitriptyline template,
and finally dried overnight under vacuum.
7.4.3 FTIR Spectroscopy
FTIR absorbance spectra were recorded on a Brüker/Vector 22 FT/IR spectrometer. The
samples were prepared according to the following procedure: the dry and purified nanoparticles
were dispersed by sonication in THF, and the solutions were subsequently spread dropwise onto
KBr pellets and allowed to dry in a desiccator.
7.4.4 Particle Size Analysis
Size analysis was performed on the nanoparticle solutions obtained after miniemulsion
polymerization and after performing approximately a 1000 times dilution to avoid interparticle
aggregation. DLS was used with a Precision Detectors PDDLS/CoolBatch+90T instrument. The
data were analyzed with the Precision Deconvolve32 Program. Measurements were taken at
163
20°C at a 90° scattering angle using a 632 nm laser source. Final sizes were obtained from the
average of at least five reproducible results.
7.4.5 AFM Imaging
AFM imaging of the nanoparticles was performed immediately after polymerization and
after resuspension of the purified particles in Millipore filtered water. The nanocapsule
concentrations were adjusted to approximately 0.1 mg/mL, and the samples were prepared by air
drying (in a desiccator for at least 24h) a few drops of the nanocapsule solutions deposited onto
freshly cleaved mica. The films were subsequently scanned in tapping mode with a Nanoscope
III AFM (Digital Instruments, Inc., Santa Barbara, CA) using silicon probes (Nanosensor
dimensions: T = 3.8-4.5 µm, W = 26-27 µm, L = 128 µm). The images were processed with a
second-order flattening routine.
7.4.6 Uptake Experiments
For binding studies, various amounts of purified and dried nanoparticles were
resuspended by sonication in 8 mL of 250 µM (amitriptyline) or 1,500 µM (bupivacaine)
HEPES-buffered saline solutions (pH = 7.4, C(HEPES) = 20 mM, C(NaCl) = 8 g/L). After
vigorous stirring for 24 h, the nanoparticles were removed by centrifugation (15 000 rpm, 1 h),
and the decrease in amitriptyline or bupivacaine concentrations in the supernatants was
quantified by UV-vis spectroscopy after preliminary establishment of calibration curves
(absorbance versus concentration) at respectively λ = 238 and 264 nm. All the data plotted result
from the average of 3 distinct uptake measurements, and the error bars correspond to the
standard deviations. The partition coefficient values (Kp) of the drugs between the nanoparticles
and the aqueous phase were estimated by dividing the number of moles of amitriptyline or
164
bupivacaine encapsulated per gram of nanoparticles by the number of moles remaining in the
aqueous solution per gram (i.e., mL) of water after uptake.
165
CHAPTER 8 CONCLUSIONS AND PERSPECTIVES
In this dissertation, the investigations on the A/W interfacial behavior of various block
copolymers in Chapters 3, 4, and 5, as well as the synthesis and characterization of nanoparticles
for drug detoxification purposes in Chapters 6 and 7 were presented.
As discussed in Chapter 3, the PS-b-PtBA dendrimer-like block copolymer aggregated into
circular surface micelles for low surface pressures before collapsing around 24 mN/m. The PS-b-
PAA dendrimer-like block copolymer formed stable Langmuir monolayers only under acidic
conditions and aggregated into circular surface micelles at low pressure, before aqueous
dissolution of the PAA segments followed by aggregation of the micellar PS cores took place
around 5 mN/m. The aggregation numbers for both dendrimer-like block copolymers were
estimated around 3-5 dendrimer-like block copolymers per circular surface micelle. These rather
low values compared to the results reported in the previous literature for other architectures and
chain lengths confirmed the tremendous influence of molecular architecture on the 2D surface
micelle formation of block copolymers.
As discussed in Chapter 4, the interfacial behavior of PEO-b-PCL block copolymers was,
similarly as for PS-b-PtBA and PS-b-PAA block copolymers, also strongly dependent on the
architecture. For the PEO-b-PCL five-arm stars, only one phase transition around 13-14 mN/m
corresponding to collapse and crystallization of the PCL segments was observed, whereas two
additional PEO-related phase transitions (aqueous dissolution at 6.5 mN/m and brush formation
at 10.5 mN/m) could be observed for the linear samples. Therefore, while the work presented in
Chapters 3 and 4 gave some hint of the interesting interfacial properties of these block
copolymers, additional investigations on other samples with different architectures and chain
lengths are necessary for a better fundamental understanding of the influence of block copolymer
166
architecture on the A/W interfacial self-assembly. Brewster angle microscopy investigations will
also be necessary on the PCL-based block copolymers in the high surface pressure region to
determine if the PCL crystals observed by AFM imaging in the LB films were formed directly at
the A/W interface or if significant additional PCL crystallization took place during transfer.
In Chapter 5, we took advantage of the surface activity of hydrosilylated PB-b-PEO three-
arm star block copolymers to synthesize at the A/W interface cross-linked polybutadiene two-
dimensional networks with PEO domains of controllable sizes trapped within, through
polycondensation of triethoxysilane pendant groups under acidic conditions. This novel and
general method to cross-link in 2D polydiene blocks should easily be applied in the future to a
variety of other polydiene-based block copolymers for the synthesis of well-defined 2D cross-
linked patterns.
In Chapter 6, to quantify the drug encapsulation selectivity of oil-filled silica nanocapsules
consisting of a hydrophobic liquid core and a silicate shell, a series of electrochemical and
spectroscopic measurements was performed, and the nanocapsules efficiently and rapidly
removed large amounts of organic compounds present in aqueous solutions. The silica shell was
shown to act similarly as a chromatographing layer, but this does not lead yet to enough
selectivity to avoid encapsulation of undesired molecules present in the blood stream. Additional
nanocapsules should therefore be designed with for instance different oil-cores or silica shells
with controllable pore sizes complementary to the target toxic drugs.
With a view toward increasing specificity in drug detoxification therapy, a series of
molecularly imprinted nanoparticles was synthesized by the non-covalent approach as described
in Chapter 7, and the encapsulation specificity was studied through rebinding experiments. These
nanoparticles are ideal candidates for in-vivo drug detoxification applications since they were
167
able to encapsulate large amounts of toxic drugs with some specificity under physiological pH
conditions. In addition to investigating their biocompatibility and biodegradability properties,
future work on the molecularly imprinted nanoparticles should primarily focus on improving
their specificity using other formulations to decrease non-specific hydrophobic binding.
168
LIST OF REFERENCES
1. (a) Gohy, J.-F. Adv. Polym. Sci. 2005, 190, 65-136. (b) Abetz, V.; Simon, P. F. W. Adv. Polym. Sci 2005, 189, 125-212. (c) Li, M.; Coenjarts, C. A.; Ober, C. K. Adv. Polym. Sci. 2005, 190, 183-226.
2. Bates, F. M.; Fredrickson, G. H. Annu. Rev. Phys. Chem. 1990, 41, 525-557.
3. Jagur-Grodzinski, J. J. Polym. Sci. Part A: Polym. Chem. 2002, 40, 2116-2133.
4. Matyjaszewski, K. Mol. Cryst. Liq. Cryst. 2004, 415, 23-34.
5. Moad, G.; Chong, Y. K.; Postma, A.; Rizzardo, E.; Thang, S. H. Polymer 2005, 46, 8458-8468.
6. Adhikari, R.; Michler, G. H.; Lebek, W.; Goerlitz, S.; Weidisch, R.; Knoll, K. J. Appl. Polym. Sci. 2002, 85, 701-713.
7. Milner, S. T. Macromolecules 1994, 27, 2333-2335.
8. Lecommandoux, S.; Borsali, R.; Schappacher, M.; Deffieux, A. ; Narayanan, T. ; Rochas, C. Macromolecules 2004 37, 1843-1848.
9. Matsen, M. W.; Bates, F. S. J. Chem. Phys. 1997, 106, 2436-2448.
10. Leibler, L. Macromolecules 1980, 13, 1602-1617.
11. Cameron, N. S.; Corbierre, M. K.; Eisenberg, A. Can. J. Chem. 1999, 77, 1311-1336.
12. Hamley, I. W. The Physics of Block Copolymers; Oxford Univ. Press, Oxford, 1998; p26.
13. Calvert, P. M. Nature 1996, 384, 311-312.
14. KSV Instruments LTD; Langmuir and Langmuir-Blodgett films, what and how?; Application note #107.
15. Vieira, J. B.; Li, Z. X.; Thomas, R. K.; Penfold, J. J. Phys. Chem. B 2002, 106, 10641-10648.
16. Dhathathreyan, A.; Baskar, G.; Ramasami, T. Langmuir 2002, 18, 4704-4708.
17. Wustneck, R.; Prescher, D.; Katholy, S.; Knochenhauer, G.; Brehmer, L. Colloids and Surfaces A: Physicochem. Eng. Aspects 2000, 175, 83-92.
18. Seo, Y-S.; Kim, K. S.; Galambos, A.; Lammertink, R. G. H.; Vancso, G. J.; Sokolov, J.; Rafailovich, M. Nano Lett. 2004, 4, 483-486.
19. Kimura, M.; Ueki, H.; Ohta, K.; Hanabusa, K.; Shirai, H.; Kobayashi, N. Langmuir 2002, 18, 7683-7687.
169
20. Currie, E. P. K.; Sieval, A. B.; Fleer, G. J.; Cohen Stuart, M. A. Langmuir 2000, 16, 8324-8333.
21. Noskov, B.A.; Lin, S.-Y.; Loglio, G.; Rubio, R. G. ; Miller, R. Langmuir 2006, 22, 2647-2652.
22. Logan, J. L.; Masse, P.; Dorvel, B.; Skolnik, A. M.; Sheiko, S. S.; Francis, R.; Taton, D.; Gnanou, Y.; Duran, R. S. Langmuir 2005, 21, 3424-3431.
23. Bowers, J.; Zarbakhsh, A.; Webster, J. R. P.; Hutchings, L. R.; Richards, R. W. Langmuir 2001, 17, 131-139.
24. Li, S.; Clarke, C. J.; Eisenberg, A.; Lennox, R. B. Thin Solid Films 1999, 354, 136-141.
25. Zhu, J.; Lennox R. B. ; Eisengerg, A. J. Phys. Chem. 1992, 96, 4727-4730.
26. Park, M.; Harrison, C.; Chaikin, P. M.; Register, R. A.; Adamson, D. H. Science 1997, 276, 1401-1404.
27. Fink, Y.; Urbas, A. M.; Bawendi, M. G.; Joannopoulos, J. D.; Thomas, E. L. J. Lightwave Technol. 1999, 17, 1963-1969.
28. Kim, S. O.; Solak, H. H.; Stoykovich, M. P.; Ferrier, N. J.; de Pablo, J. J.; Nealey, P. F. Nature 2003, 424, 411-414.
29. Dr. Don Dennis, University of Florida, Department of Anesthesiology, Query of the ASA, Closed Claims Project Database located at the University of Washington.
30. Lee, L. A.; Posner, K. L.; Domino, K. B.; Caplan, R. A.; Cheney, F. W. Anesthesiology 2004, 101, 143-152.
31. Jovanovic, A. V. Synthesis and Characterization of Oil Core Silica Shell Nanocapsules for Biomedical Applications. Ph.D. Thesis, University of Florida, Gainesville, Fl, 2006.
32. Buckley, N. A.; Dawson, A. H.; Whyte, I. M. Lancet 1994, 343, 159-162.
33. Buckley, N. A.; Whyte, I. M.; Dawson, A. H. Med. J. Aust. 1995, 164, 190-193.
34. Renehan, E. R.; Enneking, F. K.; Varshney, M.; Partch, R.; Dennis, D. M.; Morey, T. E. Reg. Anesth. Pain Med. 2005, 30, 380-384.
35. Hoffman, J. R.; Votey, S. R.; Bayer, M.; Silver, L. Am. J. Emerg. Med. 1993, 11, 336-341.
36. Bosse, G. M.; Barefoot, J. A.; Pfeifer, M. P.; Rodgers, G. C. J. Emerg. Med. 1995, 13, 203-209.
37. Weinberg, G. L.; VadeBoncouer, T.; Ramaraju, G. A.; Garcia-Amaro, M. F.; Cwik, M. J. Anesthesiology 1998, 88, 1071-1075.
170
38. Weinberg, G. L. Reg. Anesth. Pain Med. 2002, 27, 568-575.
39. Cho, H. S.; Lee, J. J.; Chung, I. S.; Shin, B. S.; Kim, J. A.; Lee, K. H. Anesth. Analg. 2000, 91, 1096-1102.
40. Ohmutra, S.; Ohta, T.; Yamamoto, K.; Kobayashi, T. A. Anesth. Analg. 1999, 88, 155-159.
41. Schenning, A. P. H. J.; Meijer, E. W. Chem. Comm. 2005, 26, 3245-3258.
42. Simon, P. F. W.; Ulrich, R.; Spiess, H. W.; Wiesner, U. Chem. Mater. 2001, 13, 3464-3486.
43. Taton, T. A. Trends Biotechnol. 2002, 20, 277-279.
44. Withesides, G. M. Nat. Biotechnol. 2003, 21, 1161-1165.
45. Cao, Y. C.; Jin, R.; Nam, J. M.; Thaxton, C. S.; Mirkin, C. A. J. Am. Chem. Soc. 2003, 125, 14676-14677.
46. Gutwein, L. G.; Webster, T. J. American Ceramic Society 26th Annual Meeting Conference Proceedings 2003.
47. Roy, I.; Ohulchanskyy, T. Y.; Pudavar, H. E.; Bergey, E. J.; Oseroff, A. R.; Morgan, J.; Dougherty, T. J.; Prasad, P. N. J. Am. Chem. Soc. 2003, 125, 7860-7865.
48. Langer, R.; Peppas, N. AIChE J. 2003, 49, 2990-3006.
49. Underhill, R. S.; Jovanovic, A. V.; Carino, S. R.; Varshney, M.; Shah, D. O.; Dennis, D. M.; Morey, T. E.; Duran, R. S. Chem. Mater. 2002, 14, 4919-4925.
50. Paul, B. K.; Moulik, S. P. Curr. Sci. 2001, 80, 990-1001.
51. Schramm, L. L.; Fisher, D. B.; Schurch, S.; Cameron, A. Colloids Surf., A 1995, 94, 145-159.
52. Gillberg, G. Emulsions and Emulsion Technology; Marcel Dekker: New York, 1984; pp 1–43.
53. Prince, L. M. Microemulsions: Theory and Practice; Academic Press, New York, 1977; p. 25.
54. Atik, S. S.; Thomas, J. K. J. Am. Chem. Soc. 1982, 104, 5868-5874.
55. Kumar, P. and Mittal, K. L. Handbook of Microemulsion Science and Technology, Marcel Dekker Inc., New York, 1999; pp. 755–771; pp. 483–497; pp. 549–603; pp. 679–712; pp. 457–482.
56. Tokuoka, Y.; Uchiyama, H.; Abe, M.; Christian, S. D. Langmuir 1995, 11, 725-729.
171
57. Staufer, C. E. Food Technol. 1992, 46, 70.
58. Aboofazeli, R.; Lawrence, M. J. Int. J. Pharm. 1993, 93, 161-175.
59. Levashov, A. V.; Khmelnitsky,Y. L.; Klyachko, N. L.; Chernyak, V. Y.; Martinek, K. J. Colloid Interface Sci. 1982, 88, 444-457.
60. Brinker, C. J.; Schere, G. W. Sol-Gel Science-The Physics and Chemistry of Sol-Gel Processing, New-York, Academic press, 1990.
61. Sol-Gel Technology; http://www.chemat.com/html/solgel.html; October 2006.
62. Wei, Y.; Xu, J. G.; Feng, Q. W.; Dong, H.; Lin, M. D. Mater. Lett. 2000, 44, 6-11.
63. Jain, T. K; Roy, I.; De, T. K.; Maitra, A. J. Am. Chem. Soc. 1998, 120, 11092-11095.
64. Agrawal, A.; Cronin, J. P.; Zhang, R. Sol. Eng. Mater. Sol. Cells 1993, 31, 9-21.
65. Wulff, G. Angew. Chem. Int. Ed. 1995, 34, 1812-1832.
66. San Vicente, B.; Navarro-Villoslada, F.; Moreno-Bondi, M. C. Anal. Bioanal. Chem. 2004, 380, 115-122.
67. Ye, L.; Haupt, K. Anal. Bioanal. Chem. 2004, 378, 1887-1897.
68. Cheng, Z.; Zhang, L.; Li, Y. Chem. Eur. J. 2004, 10, 3555-3561.
69. Pap, T.; Horvai, G. J. Chromatogr. B 2004, 804, 167-172.
70. Ikegami, K.; Lee, W. S.; Nariai, H.; Takeuchi, T. J. Chromatogr. B 2004, 804, 197-201.
71. Shim, Y. H.; Yilmaz, E.; Laviellea, S.; Haupt, K. Analyst 2004, 129, 1211-1215.
72. Tada, M.; Iwasawa, Y. J. Mol. Catal. 2003, 27, 204–205.
73. Andersson, L.; Sellergren, B.; Mosbach, K. Tetrahedron Lett. 1984, 25, 5211-5214.
74. Simon, R. L.; Spivak, D. A.; J. Chromatogr. B 2004, 804, 203-209.
75. Suedee, R.; Songkram, C.; Petmoreekul, A.; Sangkunakup, S.; Sankasa, S.; Kongyarit, N. J. Pharma. Biomed. Anal. 1999, 19, 519-527.
76. Jenkins, A. L.; Yina, R.; Jensen, J. L. Analyst, 2001, 126, 798-802.
77. Rachkov, A.; Minoura, N. J. Chromatogr. A 2000, 889, 111-118.
78. Iqbal, S. S.; Lulka, M. F.; Chambers, J. P.; Thompson, R. G.; Valdes, J. J. Mater. Sci. Eng., C 2000, 7, 77-81.
172
79. Theodoridis, G.; Konsta, G.; Bagia, C. J. Chromatogr. B 2004, 804, 43-51.
80. Huang, Y. C.; Lin, C. C.; Liu, C. Y. Electrophoresis 2004, 25, 554-561.
81. Cormack, P. A. G.; Elorza, A. Z. J. Chromatogr. B 2004, 804, 173-182.
82. Schork, F. J.; Luo, Y.; Smulders, W.; Russum, J. P.; Butté, A.; Fontenot, K. Adv. Polym. Sci. 2005, 175, 129-255.
83. Tojo, C.; Blanco, M. C.; Rivadulla, F.; López-Quintela, M. A. Langmuir 1997, 13, 1970-1977.
84. Landfester, K.; Tiarks, F.; Hentze, H. P.; Antonietti, M. Macromol. Chem. Phys. 2000, 201, 1-5.
85. Barrère, M.; Landfester, K. Polymer 2003, 44, 2833-2841.
86. Marie, E.; Rothe, R.; Antonietti, M.; Landfester, K. Macromolecules 2003, 36, 3967-3973.
87. Roberts, G. Langmuir-Blodgett films; Plenum Press: New-York, 1990.
88. Schwartz, D. K. Surface Science Reports 1997, 27, 245-334.
89. Digital instruments Nanoscope III Multimode Scanning Probe Microscope Instruction Manual; 1993.
90. Electron Microscopy; http://www.unl.edu/CMRAcfem/em.htm; October 2006.
91. Dynamic Light Scattering, http://www.bic.com/DLSBasics.html; October 2006.
92. Cyclic Voltammetry Primer; http://www-biol.paisley.ac.uk/marco/Enzyme_Electrode/Chapter1/Cyclic_Voltammetry1.htm#cyclic; October 2006.
93. (a) Hadjichristidis, N. J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 857-871. (b) Yang, L.; Hong, S.; Gido, S. P.; Velis, G.; Hadjichristidis, N. Macromolecules 2001, 34, 9069-9073. (c) Zhu, L.; Huang, P.; Chen, W. Y.; Ge, Q.; Quirk, R. P.; Cheng, S. Z. D.; Thomas, E. L.; Lotz, B.; Hsiao, B. S.; Yeh, F.; Liu, L. Macromolecules 2002, 35, 3553-3562.
94. (a) Pispas, S.; Poulos, Y.; Hadjichristidis, N. Macromolecules 1998, 31, 4177-4181. (b) Pispas, S.; Hadjichristidis, N.; Potemkin, I.; Khokhlov, A. Macromolecules 2000, 33, 1741-1746.
173
95. (a) Nakano, M.; Deguchi, M.; Endo, H.; Matsumoto, K.; Matsuoka, H.; Yamaoka, H. Macromolecules 1999, 32, 6088-6092. (b) Cox, J. K.; Eisenberg, A.; Lennox, R. B. Curr. Opin. Colloid Interface Sci. 1999, 4, 52-59. (c) Zhao, B.; Brittain, W. J. Prog. Polym. Sci. 2000, 25, 677-710. (d) Tsukruk, V. V.; Sidorenko, A.; Gorbunov, B. B.; Chizhik, S. A. Langmuir 2001, 17, 6715-6719. (e) Mahltig, B.; Jérôme, R.; Stamm, M. Phys. Chem. Chem. Phys. 2001, 3, 4371-4375. (f) Anastasiadis, S. H.; Retsos, H.; Pispas, S.; Hadjichristidis, N.; Neophytides, S. Macromolecules 2003, 36, 1994-1999.
96. (a) Cox, J. K.; Yu, K.; Eisenberg, A.; Lennox, R. B. Phys. Chem. Chem. Phys. 1999, 1, 4417-4421. (b) Cox, J. K.; Yu, K.; Constantine, B.; Eisenberg, A.; Lennox, R. B. Langmuir 1999, 15, 7714-7718. (c) Baker, S. M.; Leach, K. A.; Devereaux, C. E.; Gragson, D. E. Macromolecules 2000, 33, 5432-5436. (d) Devereaux, C. A.; Baker, S. M. Macromolecules 2002, 35, 1921-1927. (e) Francis, R.; Skolnik, A. M.; Carino, S. R.; Logan, J. L.; Underhill, R. S.; Angot, S.; Taton, D.; Gnanou, Y.; Duran, R. S. Macromolecules 2002, 35, 6483-6485. (f) Peleshanko, S.; Jeong, J.; Gunawidjaja, R.; Tsukruk, V. V. Macromolecules 2004, 37, 6511-6522. (g) Cheyne, R. B.; Moffit, M. G. Langmuir 2005, 21, 5453-5460. (h) Kim, Y.; Pyun, J.; Fréchet, J. M. J.; Hawker, C. J.; Frank, C. W. Langmuir 2005, 21, 10444-10458. (i) Gunawidjaja, R.; Peleshanko, S.; Tsukruk, V. V. Macromolecules 2005, 38, 8765-8774. (j) Gunawidjaja, R.; Peleshanko, S.; Genson, K. L.; Tsitsilianis, C.; Tsukruk, V. V. Langmuir 2006, 22, 6168-6176.
97. Li, S.; Hanley, S.; Khan, I.; Varshney, S. K.; Eisenberg, A.; Lennox, R. B. Langmuir 1993, 9, 2243-2246.
98. (a) Niwa, M.; Katsurada, N.; Higashi, N. Macromolecules 1988, 21, 1880-1883. (b) Niwa, M.; Hayashi, T.; Higashi, N. Langmuir 1990, 6, 263-268. (c) Meszaros, M.; Eisenberg, A.; Lennox, R. B. Faraday Discuss. 1994, 98, 283-294. (d) Currie, E. P. K.; Sieval, A. B.; Avena, M.; Zuilhof, H.; Sudhölter, E. J. R.; Cohen Stuart M. A. Langmuir 1999, 15, 7116-7118.
99. Genson, K. L.; Hoffman, J.; Teng, J.; Zubarev, E. R.; Vaknin, D.; Tsukruk, V. V. Langmuir 2004, 20, 9044-9052.
100. Matmour, R.; Lepoittevin, B.; Joncheray, T. J.; El-Khouri, R. J.; Taton, D.; Duran, R. S.; Gnanou, Y. Macromolecules 2005, 38, 5459-5467.
101. Li, B.; Wu, Y.; Liu, M. ; Esker, A. R. Langmuir 2006, 22, 4902-4905.
102. (a) Sutherland, J. E.; Miller, M. L. J. Polym. Sci. B Polym. Lett. 1969, 7, 871-876. (b) Mudgil, P.; Dennis, G. R.; Millar, T. J. Langmuir 2006, 22, 7672-7677.
103. Logan, J. L.; Masse, P.; Gnanou, Y.; Taton, D.; Duran, R. S. Langmuir 2005, 21, 7380-7389.
104. Mengel, C.; Esker, A. R.; Meyer, W. H.; Wegner, G. Langmuir 2002, 18, 6365-6372.
105. Zhu, J.; Lennox, R. B.; Eisenberg, A. Langmuir 1991, 7, 1579-1584.
174
106. Zeng, H.; Maeda, N.; Chen, N.; Tirrel, M.; Israelachvili, J. Macromolecules 2006, 39, 2350-2363.
107. Crisp, D. J. J. Coll. Sci. 1946, 1, 49-70.
108. (a) Mori, H.; Chan Seng, D.; Lechner, H.; Zhang, M.; Müller, A. H. E. Macromolecules 2002, 35, 9270-9281. (b) Sonnenberg, L.; Parvole, J.; Borisov, O.; Billon, L.; Gaub, H. E.; Seitz, M. Macromolecules 2006, 39, 281-288.
109. (a) Ishimuro, Y.; Ueberreiter, K. Colloid. Polym. Sci. 1980, 258, 928-931. (b) Okubo, T. J. Colloid Interface Sci. 1988, 125, 386-398.
110. Gonçalves da Silva, A. M.; Filipe, E. J. M.; d’Oliveira, J. M. R.; Martinho, J. M. G Langmuir 1996, 12, 6547-6553.
111. Gan, Z.; Jim, T. F.; Li, M.; Yuer, Z.; Wang, S.; Wu, C. Macromolecules 1999, 32, 590-594.
112. Crisci, L.; Della Volpe, C.; Maglio, G.; Nese, G.; Palumbo, R.; Rachiero, G. P.; Vignola, M. C. Macromol. Biosci. 2003, 3, 749-757.
113. Choi, Y. K.; Bae, Y. H.; Kim, S. W. Macromolecules 1998, 31, 8766-8774.
114. Deng, M.; Chen, X.; Piao, L.; Zhang, X.; Dai, Z.; Jing, X. J. Polym. Sci., Part A: Polym. Chem. 2004, 42(4), 950-959.
115. Allen, C.; Han, J.; Yu, Y.; Maysinger, D.; Eisenberg, A. J. Controlled Release 2000, 63, 275-286.
116. Storck, M.; Orend, K. H.; Schmitzrixen, T. J. Vasc. Surg. 1993, 27, 413–424.
117. Kuzmenka, D. J.; Granick, S. Macromolecules 1988, 21, 779-782.
118. Leiva, A.; Gargallo, L.; Radic, D. J. Macromol. Sci. Part A: Pure Appl. Chem. 2004, 41(5), 577-583.
119. Li, B.; Esker, A. R. Master’s Thesis, etd-09232004-105021, 2004.
120. Lee, I. C.; Frank, C. W.; Wursch, A.; Hedrick, J. L. Polym. Prepr. 2005, 46(1), 494-495.
121. (a) Meier, M. A. R.; Gohy, J.-F.; Fustin, C.-A.; Schubert, U. S. J. Am. Chem. Soc. 2004, 126, 11517-11521. (b) Meier, M. A. R.; Aerts, S. N. H.; Staal, B. B. P.; Rasa, M.; Schubert, U. S. Macromol. Rapid Commun. 2005, 26, 1918-1924.
122. Glass, J. E. J. Phys. Chem. 1968, 72, 4459-4467.
123. Massa, M. V.; Dalnoki-Veress, K.; Forrest, J. A. Eur. Phys. J. E 2003, 11, 191-198.
124. Ok, S.; Levent Demirel, A. J. Macromol. Sci., Part B: Phys. 2003, B42 (3&4), 611-620.
175
125. Huang, L.; Nishinari, K. J. Polym. Sci. Part B: Polym. Phys. 2001, 39(5), 496-506.
126. Cohn, D.; Stern, T.; Gonzalez, M. F.; Epstein, J. J. Biomed. Mater. Res. 2002, 59(2), 273-281.
127. Ivanova, A.; Tadjer, A.; Radoev, B. Int. J. Quantum. Chem. 2002, 89, 397-404.
128. Lee, W-K.; Gardella, J. A. Langmuir 2000, 16, 3401-3406.
129. Lee, W-K.; Nowak, R. W.; Gardella, J. A. Langmuir 2002, 18, 2309-2312.
130. Adams, J.; Buske, A.; Duran, R. S. Macromolecules 1993, 26, 2871-2877.
131. Heck, B.; Sadiku, E. R.; Strobl, G. R. Macromol. Symp. 2001, 165, 99-114.
132. Berrill, S. A.; Heatley, F.; Collett, J. H.; Attwood, D.; Booth, C.; Fairclough, J. P. A.; Ryan, A. J.; Viras, K.; Dutton, A. J.; Blundell, R. S. J. Mater. Chem. 1999, 9, 1059-1063.
133. Iwata, T.; Furuhashi, Y.; Su, F.; Doi, Y. Riken Rev. 2001, 42, 15-18.
134. Mareau, V. H.; Prud’homme, R. E. Macromolecules 2005, 38, 398-408.
135. Qiao, C.; Zhao, J.; Jiang, S.; Ji. X.; An, L.; Jiang, B. J. Polym. Sci. Part B: Polym. Phys. 2005, 43(11), 1303-1309.
136. Kressler, J.; Wang, C.; Krammer, H. W. Langmuir 1997, 13, 4407-4412.
137. Gan, Z.; Zhang, J.; Jiang, B. J. Appl. Polym. Sci. 1997, 63(13), 1793-1804.
138. Shiomi, T.; Imai, K.; Takenaka, K.; Takeshita, H.; Hayashi, H.; Tezuka, Y. Polymer 2001, 42, 3233-3239.
139. He, C.; Sun, J.; Deng, C.; Zhao, T.; Deng, M.; Chen, X.; Jing, X. Biomacromolecules 2004, 5, 2042-2047.
140. Jiang, S.; He, C.; An, L.; Chen, X.; Jiang, B. Macromol. Chem. Phys. 2004, 205, 2229-2234.
141. Li, S. M.; Chen, X. H.; Gross, R. A.; McCarthy, S. P. J. Mater. Sci.: Mater. Med. 2000, 11, 227-233.
142. Bittiger, H.; Marchessault, R. H., Niegisch, W. D. Acta Crystallogr. 1970, B26, 1923-1927.
143. Holland, N. B.; Xu, Z.; Vacheethasanee, K.; Marchant R. E. Macromolecules 2001, 34, 6424-6430.
144. Fauré, M. C.; Bassereau, P.; Lee, L. T.; Menelle, A.; Lheveder, C. Macromolecules 1999, 32, 8538-8550.
176
145. Fauré, M. C.; Bassereau, P.; Carignano, M. A.; Szleifer, I.; Gallot, Y.; Andelman, D. Eur. Phys. J. B 1998, 3, 355-375.
146. Hottle, J. R.; Deng, J.; Kim, H-J.; Farmer-Creely, C. E.; Viers, B. D.; Esker, A. R. Langmuir 2005, 21, 2250-2259.
147. Hottle, J. R.; Kim, H.-J.; Deng, J.; Farmer-Creely, C. E.; Viers, B. D.; Esker, A. R. Macromolecules 2004, 37, 4900-4908.
148. Malzert, A.; Boury, F.; Saulnier, P.; Benoit, J. P.; Proust, J. E. Langmuir 2001, 17, 7837-7841.
149. Malzert, A.; Boury, F.; Saulnier, P.; Benoit, J. P.; Proust, J. E. Langmuir 2000, 16, 1861-1867.
150. Kim, H.-J.; Deng, J.; Hoyt Lalli, J.; Riffle, J. S.; Viers, B. D.; Esker, A. R. Langmuir 2005, 21, 1908-1916.
151. Nagata, K.; Kawaguchi, M. Macromolecules 1990, 23, 3957-3962.
152. Kawaguchi, M.; Nishida, R. Langmuir 1990, 6, 492-496.
153. Peleshanko, S.; Gunawidjaja, R.; Jeong, J.; Shevchenko, V. V.; Tsukruk, V. V. Langmuir 2004, 20, 9423-9427.
154. Li, Z.; Zhao, W.; Quinn, J.; Rafailovich, M. H.; Sokolov, J.; Lennox, R. B.; Eisenberg, A.; Wu, X. Z.; Kim, M. W.; Sinha, S. K.; Tolan, M. Langmuir 1995, 11, 4785-4792.
155. Kuo, S.-W.; Lin, C.-L.; Chang, F.-C. Macromolecules 2002, 35, 278-285.
156. Gonçalves da Silva, A. M.; Simões Gamboa, A. L.; Martinho, J. M. G. Langmuir 1998, 14, 5327-5330.
157. Wesemann, A.; Ahrens, H.; Förster, S.; Helm, C. A. Langmuir 2004, 20, 11528-11535.
158. Rippner, B.; Boschkova, K.; Claesson, P. M.; Arnebrant, T. Langmuir 2002, 18, 5213-5221.
159. Barentin, C.; Muller, P.; Joanny, J. F. Macromolecules 1998, 31, 2198-2211.
160. O’Connor, S. M.; Gehrke, S. H.; Retzinger, G. S. Langmuir 1999, 15, 2580-2585.
161. Dorn, K.; Klingbiel, R. T.; Specht, D. P.; Tyminski, P. N.; Ringsdorf, H.; O'Brien, D. F. J. Am. Chem. Soc. 1984, 106, 1627-1633.
162. Regen, S. L.; Czech, B.; Singh, A. J. Am. Chem. Soc. 1980, 102, 6638-6640.
163. Won, Y. Y.; Davis, H. T.; Bates, F. S. Science 1999, 283, 960-963.
177
164. Liu, G. J.; Qiao, L. J.; Guo, A. Macromolecules 1996, 29, 5508-5510.
165. Thurn-Albrecht, T.; Schotter, J.; Kästle, G. A.; Emley, N.; Shibauchi, T.; Krusin-Elbaum, L.; Guarini, K.; Black, C. T.; Tuominen, M. T.; Russell, T. P. Science 2000, 290, 2126-2129.
166. Doshi, D. A.; Huesing, N. K.; Lu, M.; Fan, H.; Lu, Y.; Simmons-Potter, K.; Potter, B. G., Jr.; Hurd, A. J.; Brinker, C. J. Science 2000, 290, 107-111.
167. Discher, B. M.; Bermudez, H.; Hammer, D. A.; Discher, D. E.; Won, Y. Y.; Bates, F. S. J. Phys. Chem. B 2002, 106, 2848-2854.
168. Discher, D. E.; Won, Y. Y.; Ege, D. S.; Lee, J. C. M.; Bates, F. S.; Discher, B. M.; Hammer, D. A. Science 1999, 284, 1143-1146.
169. Bubeck, C. Thin Solid Films 1988, 160, 1-14.
170. Kloeppner, L. J.; Duran, R. S. Langmuir 1998, 14, 6734-6742.
171. Kloeppner, L. J.; Duran, R. S. J. Am. Chem. Soc. 1999, 121, 8108-8109.
172. Kloeppner, L. J.; Batten, J. H.; Duran, R. S. Macromolecules 2000, 33, 8006-8011.
173. Bruno, F. F.; Akkara, J. A.; Samuelson, L. A.; Kaplan, D. L.; Mandal, B. K.; Marx, K. A.; Kumar, J.; Tripathy, S. K. Langmuir 1995, 11, 889-892.
174. Kimkes, P.; Sohling, U.; Oostergetel, G. T.; Schouten, A. J. Langmuir 1996, 12, 3945-3951.
175. Fichet, O.; Tran-Van, F.; Teyssie, D.; Chevrot, C. Thin Solid Films 2002, 411, 280-288.
176. Fichet, O.; Plesse, C.; Teyssie, D. Colloids Surf., A: Physicochem. Eng. Aspects 2004, 244, 121-130.
177. Carino, S. R.; Totsmann, H.; Underhill, R.; Logan, J.; Weerasekera, G.; Culp, J.; Davidson, M.; Duran, R. S. J. Am. Chem. Soc. 2001, 123, 767-768.
178. Carino, S. R.; Duran, R. S.; Baney, R. H.; Gower, L. A.; He, L.; Sheth, P. K. J. Am. Chem. Soc. 2001, 123, 2103-2104.
179. Carino, S. R.; Duran, R. S. Macromol. Chem. Phys. 2005, 206, 83-94.
180. Rehage, H.; Veyssié, M. Angew. Chem., Int. Ed. Engl. 1990, 29, 439-448.
181. Sisson, T. M.; Lamparski, H. G.; Kölchens, S.; Elayadi, A.; O'Brien, D. F. Macromolecules 1996, 29, 8321-8329.
182. Fichet, O.; Teyssié, D. Macromolecules 2002, 35, 5352-5354.
178
183. Markowitz, M. A.; Bielski, R.; Regen, S. L. J. Am. Chem. Soc. 1988, 110, 7545-7546.
184. Markowitz, M. A.; Janout, V.; Castner, D. G.; Regen, S. L. J. Am. Chem. Soc. 1989, 111, 8192-8200.
185. Conner, M.; Janout, V.; Regen, S. L. J. Am. Chem. Soc. 1993, 115, 1178-1180.
186. Conner, M. D.; Janout, V.; Kudelka, I.; Dedek, P.; Zhu, J.; Regen, S. L. Langmuir 1993, 9, 2389-2397.
187. Dedek, P.; Webber, A. S.; Janout, V.; Hendel, R. A.; Regen, S. L. Langmuir 1994, 10, 3943-3945.
188. Lee, W.; Hendel, R. A.; Dedek, P.; Janout, V.; Regen, S. L. J. Am. Chem. Soc. 1995, 117, 6793-6794.
189. Zhang, L.-H.; Hendel, R. A.; Cozzi, P.; Regen, S. L. J. Am. Chem. Soc. 1999, 121, 1621-1622.
190. Dubault, A.; Veyssié, M.; Liebert, L.; Strzelecki, L. Nature, Phys. Sci. 1973, 245, 94-95.
191. Dubault, A.; Casagrande, C.; Veyssié, M. J. Phys. Chem. 1975, 79, 2254-2259.
192. Rehage, H.; Veyssié, M. Angew. Chem. 1990, 102, 497-506.
193. Rehage, H.; Schnabel, E.; Veyssié, M. Makromol. Chem. 1988, 189, 2395-2408.
194. Harrison, R. M.; Brotin, T.; Noll, B. C.; Michl, J. Organometallics 1997, 16, 3401-3412.
195. Michl, J.; Magnera, T. F. PNAS 2002, 99, 4788-4792.
196. Palacin, S.; Barraud, A. Colloids Surf. 1991, 52, 123-147.
197. Porteu, F.; Palacin, S.; Ruaudel-Teixier, A.; Barraud, A. J. Phys. Chem. 1991, 95, 7438-7447.
198. Palacin, S.; Porteu, F.; Ruaudel-Teixier, A. Thin Films 1995, 20, 69-82.
199. Magnera, T. F.; Pesherbe, L. M.; Körblova, E.; Michl, J. J. Organomet. Chem. 1997, 548, 83-89.
200. Brotin, T.; Pospisil, L.; Fiedler, J.; HKing, B. T.; Michl, J. J. Phys. Chem. B. 1998, 102, 10062-10070.
201. Magnera, T. F.; Michl, J. Actual. Fis. Quim. Org. 1998, 50-55.
202. Pospisil, L.; Heyrovsky, M.; Pecka, J.; Michl, J. Langmuir 1997, 13, 6294-6301.
203. Francis, R.; Louche, G.; Duran, R. S. Thin Solid Films 2006, 513, 347-355.
179
204. (a) Britt, D. W.; Hlady, V. J. Phys. Chem. B 1999, 103, 2749-2754. (b) Vidon, S.; Leblanc, R. M. J. Phys. Chem. B 1998, 102, 1279-1286. (c) Sjöblom, J.; Stakkestad, G.; Ebeltoft, H.; Friberg, S. E.; Claesson, P. Langmuir 1995, 11, 2652-2660. (d) Brousseau, J.-L.; Vidon, S.; Leblanc, R. M. J. Chem. Phys. 1998, 108, 7391-7396. (e) Britt, D. W.; Hlady, V. Langmuir 1999, 15, 1770-1776. (f) Sjöblom, J.; Ebeltoft, H.; Bjrseth, A.; Friberg, S. E.; Brancewicz, C. J. J. Dispersion Sci. Technol. 1994, 15, 21-34. (g) Peters, R. A. Trans. Faraday Soc. 1930, 26, 197. (h) Seeboth, A.; Hettrich, W. J. Adhes. Sci. Technol. 1997, 11, 495-505. (i) Chen, C.; Loch, C. L.; Wang, J.; Chen, Z. J. Phys. Chem. B 2003, 107, 10440-10445. (j) Linden, M.; Slotte, J. P.; Rosenholm, J. B. Langmuir 1996, 12, 4449-4454. (k) Fontaine, P.; Goldmann, M.; Rondelez, F. Langmuir 1999, 15, 1348-1352. l) Park, H. K.; Ha, T. H.; Kim, K. Langmuir, 2004, 20, 4851-4858.
205. Brooks, C. F.; Thiele, J.; Frank, C. W.; O'Brien, D. F.; Knoll, W.; Fuller, G. G.; Robertson, C. R. Langmuir 2002, 18, 2166-2173.
206. Guo, X.; Farwaha, R.; Rempel, G. L. Macromolecules 1990, 23, 5047-5054.
207. Guo, X.; Rempel, G. L. Macromolecules 1992, 25, 883-886.
208. Gabor, A. H.; Lehner, E. A.; Mao, G.; Schneggenburger, L. A.; Ober, C. K. Chem. Mater. 1994, 6, 927-934.
209. Witte, J.; Guenter, L.; Pampus, G.: (to Bayer A. G.) Ger. Offen. 2,344,734, 1975.
210. Kimitake, K.: (to Shin-Etsu Chemical Industry Co.) Japan Kokai 7,777,194, 1977.
211. Voigt, H.; Winter, R.: (to Kahel und Metallwerke Gutehoffnungshutte) Germ. Offen. 2,646,080, 1978.
212. Uemiya, T.; Osawa, Y.; Shibita, Y.; Nishimura, A.; Niwa, S.: (to Sumitomo Electric Industries, Ltd.) Japan Kokai 61,254,625, 1986.
213. Lien, O. S.; Humphreys, R. W.: (to Loctite Corp.) U. S. Patent 4,587,276, 1986.
214. Fontanille, M.; Krantz, N.; Gautier, C.; Raynal, S.: (Société Nationale des Poudres et Explosifs) Fr. Demande FR 2,579,982, 1986.
215. Voigt, H.; Muller, G.: (to Kahel und Metallwerke Gutehoffnungshutte A. G.) Germ. Offen. 2,554,944, 1977.
216. Tsai, T.: (to Copolymer Rubber and Chemical Corp.) U.S. Patent 4,158,765, 1979.
217. Streck, R.; Zerpner, D.; Haag, H.; Nordsick, K. H.: (to Chemistry Werke Huels A. G.) Germ. Offen. 2,635,601, 1979.
218. Hempenius, M. A.; Michelberger, W.; Möller, M. Macromolecules 1997, 30, 5602-5605.
219. Xenidou, M.; Hadjichristidis, N. Macromolecules 1998, 31, 5690-5694.
180
220. Baum, K.; Baum, J. C.; Ho, T. J. Am. Chem. Soc. 1998, 120, 2993-2996.
221. Iraqi, A.; Seth, S.; Vincent, C. A.; Cole-Hamilton, D. J.; Watkinson, M. D.; Graham, I. M.; Jeffrey, D. J. Mater. Chem. 1992, 2, 1057-1064.
222. Matmour, R.; Francis, R.; Duran, R. S.; Gnanou, Y. Macromolecules 2005, 38, 7754-7767.
223. Kawaguchi, M.; Tohyama, M.; Mutoh, Y.; Takahashi, A. Langmuir 1988, 4, 407-410.
224. Henderson, J. A.; Richards, R. W.; Penfold, J.; Thomas, R. K.; Lu, J. R. Macromolecules 1993, 26, 4591-4600.
225. Gago, R.; Vasquez, L.; Cuerno, R.; Varela, M.; Ballesteros, C.; Albella, J. M. Appl. Phys. Lett. 2001, 78, 3316-3318.
226. Jovanovic, A. V.; Underhill, S. R.; Bucholz, T. L. and Duran, R. S. Chem. Mater. 2005, 17, 3375-3383.
227. Wissing, S. A.; Muller, R. H. J. Controlled Release 2002, 81, 225-233.
228. Zara, G. P.; Bargoni, A.; Cavalli, R.; Fundaro, A.; Vighetto, D.; Gasco, M. R. J. Pharm. Sci 2002, 91, 1324-1333.
229. Kreuter, J. J. Controlled Release 1991, 16, 169-176.
230. Brigger, I.; Chaminade, P.; Marsaud, V.; Appel, M.; Besnard, M.; Gurny, R.; Renoir, M.; Couvreur, P. Int. J. Pharm. 2001, 214, 37-42.
231. Li, Y. P.; Pei, Y. Y.; Zhang, X. Y.; Gu, Z.; Zhou, Z. H.; Yuan, W. F.; Zhou, J. J.; Zhu, J. H.; Gao, X. J. J. Controlled Release 2001, 71, 203-211.
232. Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Chem. Rev. 1999, 99, 3181-3198.
233. Ouchi, T.; Toyohara, M.; Arimura, H.; Ohya, Y. Biomacromolecules 2002, 3, 885-888.
234. Ulrich, A. S. Biosci. Rep. 2002, 22, 129-150.
235. Liu, S. C.; O’Brien, D. F. J. Am. Chem. Soc. 2002, 124, 6037-6042.
236. Discher, D. E.; Eisenberg, A. Science 2002, 297, 967-973.
237. Orr, G. W.; Barbour, L. J.; Atwood, J. L. Science 1999, 285, 1049-1052.
238. Bartoli, S.; Roelens, S. J. Am. Chem. Soc. 2002, 124, 8307-8315.
239. Zhang, Q.; Remsen, E. E.; Wooley, K. L. J. Am. Chem. Soc. 2000, 122, 3642-3651.
240. Wooley, K. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 1397-1407.
181
241. Gálvez, A. M.; Mateo García, J. V.; Martínez Calatayud, J. J. Pharm. Biomed. Anal. 2002, 30, 535-545.
242. Wiberg, K. Ph.D. Thesis, Stockholm University, 2004; Chapter 3.
243. Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568.
244. Hickman, J. J.; Ofer, D.; Laibinis, P. E.; Whitesides, G. M.; Wrighton, M. S. Science 1991, 252, 688-691.
245. Stone, D. L.; Smith, D. K.; McGrail, P. T. J. Am. Chem. Soc. 2002, 124, 856-864.
246. Pollak, K. W.; Leon, J. W.; Fréchet, J. M. J.; Maskus, M.; Abruňa, H. D. Chem. Mater. 1998, 10, 30-38.
247. Mandal, A. B.; Nair, B. U.; Ramaswamy, D. Langmuir 1988, 4, 736-739.
248. Mandal, A. B.; Nair, B. U. J. Phys. Chem. 1991, 95, 9008-9013.
249. Mandal, A. B. Langmuir 1993, 9, 1932-1933.
250. Asakawa, T.; Sunagawa, H.; Miyagishi, S. Langmuir 1998, 14, 7091-7094.
251. Howells, A. R.; Zambrano, P. J.; Collinson, M. M. Anal. Chem. 2000, 72, 5265-5271.
252. Bard, A. J.; Faulkner, L. K. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley and Sons: New York, 2001; Chapter 4.
253. Litovitz, T. Drug Saf. 1998, 18, 9-19.
254. Fallon, M. S.; Varshney, M.; Dennis, D. M.; Chauhan, A. J. Pharmacokin. Pharmacodyn. 2004, 31, 381-400.
255. (a) Weinberg, G.; Ripper, R.; Feinstein, D. L.; Hoffman, W. Reg. Anesth. Pain Med. 2003, 28, 198-202. (b)/ Morey, T. E.; Varshney, M.; Flint, J. A.; Rajasekaran, S.; Shah, D. O.; Dennis, D. M. Nano Lett. 2004, 4, 757-759. (c) Morey, T. E.; Varshney, M.; Flint, J. A.; Seubert, C. N.; Smith, W. B.; Bjoraker, D. G.; Shah, D. O.; Denis, D. M. J. Nanopart. Res. 2004, 6, 159-170. (d) Varshney, M.; Morey, T. E.; Shah, D. O.; Flint, J. A.; Moudgil, B. M.; Seubert, C. N.; Dennis, D. M. J. Am. Chem. Soc. 2004, 126, 5108-5112.
256. (a) Lee, D.-W.; Baney, R. H. Biomacromolecules 2004, 5, 1310-1315. (b) Lee, D.-W.; Flint, J.; Morey, T.; Dennis, D.; Partch, R.; Baney, R. J. Pharm. Sci. 2005, 94, 373-381. (c) Lee, D.-W.; Baney, R. Biotechnol. Lett. 2004, 26, 713-716. (d) Lee, D.-W.; Powers, K.; Baney, R. Carbohydr. Polym. 2004, 58, 371-377.
257. (a) Arshady, R.; Mosbach, K. Makromol. Chem. 1981, 182, 687-692. (b) Brüggemann, O.; Haupt, K.; Ye, L.; Yilmaz, E.; Mosbach, K. J. Chromatogr. A 2000, 889, 14-24.
182
258. (a) Khasawneh, M. A.; Vallano, P. T.; Remcho, V. T. J. Chromatogr. A 2001, 922, 87-97. (b) Vallano, P. T.; Remcho, V. T. J. Chromatogr. A 2000, 888, 23-34. (c) Schweitz, L.; Andersson, L. I.; Nilsson, S. J. Chromatogr. A 1997, 792, 401-409.
259. (a) Ki, C. D.; Chang, J. Y. Macromolecules 2006, 39, 3415-3419. (b) Pérez, N.; Withcombe, M. J.; Vulfson, E. N. J. Appl. Polym. Sci. 2000, 77, 1851-1859. (c) Carter, S. R.; Rimmer, S. Adv. Mater. 2002, 14, 667-670. (d) Carter, S.; Lu, S.-Y.; Rimmer, S. Supramol. Chem. 2003, 15, 213-220. (e) Carter, S. R.; Rimmer, S. Adv. Funct. Mater. 2004, 14, 553-561. (f) Markowitz, M. A.; Kust, P. R.; Klaehn, J.; Deng, G.; Gaber, B. P. Analytica Chimica Acta 2001, 435, 177-185. (g) Pérez-Moral, N.; Mayes, A. G. Anal. Chim. Acta 2004, 504, 15-21. (h) Tovar, G. E. M.; Kräuter, I.; Gruber, C. Top. Curr. Chem. 2003, 227, 125-144. (i) Ye, L.; Weiss, R.; Mosbach, K. Macromolecules 2000, 33, 8239-8245. (j) Li, Z.; Ding, J.; Day, M.; Tao, Y. Macromolecules 2006, 39, 2629-2636.
260. Schork, F. J.; Luo, Y.; Smulders, W.; Russum, J. P.; Butté, A.; Fontenot, K. Adv. Polym. Sci. 2005, 175, 129-255.
261. (a) Vaihinger, D.; Landfester, K.; Kräuter, I.; Brunner, H.; Tovar, G. E. M. Macromol. Chem. Phys. 2002, 203, 1965-1973. (b) Lehmann, M.; Dettling, M.; Brunner, H.; Tovar, G. E. M. J. Chromatogr. B 2004, 808, 43-50. (c) Weber A.; Dettling, M.; Brunner, H.; Tovar, G. E. M. Macromol. Rapid Commun. 2002, 23, 824-828.
262. Antonietti, M.; Lohmann, S.; Van Niel, C. Macromolecules 1992, 25, 1139-1143.
263. Schwarz, L.; Holdsworth, C. I.; McCluskey, A.; Bowyer, M. C. Aust. J. Chem. 2004, 57, 759-764.
264. (a) Austin, R. P.; Barton, P.; Cockroft, S. L.; Wenlock, M. C.; Riley, R. J. Drug Met. Disp. 2002, 30, 1497-1503. (b) Li, J. Chromatographia 2005, 61, 479-492.
265. (a) Park, D. W.; Haam, S.; Lee, T. G.; Kim, H.-S.; Kim, W.-S. J. Biomed. Mat. Res., Part A 2004, 71, 497-507 (b) Ma, Z.-Y.; Guan, Y.-P.; Liu, H.-Z. Polym. Int. 2005, 54, 1502-1507. (c) Lee, T. Y.; Kaung, W.; Jönsson, E. S.; Lowery, K.; Guymon, C. A.; Hoyle, C. E. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 4424-4436.
266. (a) Dong, W.; Yan, M.; Zhang, M.; Liu, Z.; Li, Y. Anal. Chim. Acta 2005, 542, 186-192. (b) Li, Z.; Day, M.; Ding, J.; Faid, K. Macromolecules 2005, 38, 2620-2625. (c) Rufino, E. S.; Monteiro, E. E. C. Polymer 2003, 44, 7189-7198. (d) Wang, H. Y.; Xia, S. L.; Sun, H.; Liu, Y. K.; Cao, S. K.; Kobayashi, T. J. Chromatogr. B 2004, 804, 127-134. (e) Syu, M.-J.; Deng, J.-H.; Nian, Y.-M. Anal. Chim. Acta 2004, 504, 167-177.
183
267. (a) Karlsson, J. G.; Karlsson, B.; Andersson, L. I.; Nicholls, I. A. Analyst 2004, 129, 456-462. (b) Andersson, L. I.; Hardenborg, E.; Sandberg-Ställ, M.; Möller, K.; Henriksson, J.; Bramsby-Sjöström, I.; Olsson, L.-I.; Abdel-Rehim, M. Anal. Chim. Acta 2004, 526, 147-154. (c) Dirion, B.; Cobb, Z.; Schillinger, E.; Andersson, L. I.; Sellergren, B. J. Am. Chem. Soc. 2003, 125, 15101-15109. (d) Karlsson, J. G.; Andersson, L. I.; Nicholls, I. A. Anal. Chim. Acta 2001, 435, 57-64. (e) Zheng, N.; Li, Y.-Z.; Chang, W.-B.; Wang, Z.-M.; Li, T.-J. Anal. Chim. Acta 2002, 452, 277-283. (f) Piletsky, S. A.; Andersson, H. S.; Nicholls, I. A. J. Mol. Recogn. 1998, 11, 94-97. (g) Yu, C.; Mosbach, K. J. Mol. Recogn. 1998, 11, 69-74. (h) Piletsky, S. A.; Andersson, H. S.; Nicholls, I. A. Macromolecules 1999, 32, 633-636. (i) Gore, M. A.; Karmalkar, R. N.; Kulkarni, M. G. J. Chromatogr. B 2004, 804, 211-221. (j) Haginaka, J.; Kagawa, C. Anal. Bioanal. Chem. 2004, 378, 1907-1912.
268. Chapuis, F.; Pichon, V.; Lanza, F.; Sellergren, B.; Hennion, M.-C. J. Chromatogr. B 2004, 804, 93-101.
269. Huang, X.; Kong, L.; Li, X.; Zheng, C.; Zou, H. J. Mol. Recogn. 2003, 16, 406-411.
270. Jovanovic, A. V.; Flint, J. A.; Varshney, M.; Morey, T. E.; Dennis, D. M.; Duran, R. S. Biomacromolecules 2006, 7, 945-949.
184
BIOGRAPHICAL SKETCH
Thomas J. Joncheray was born in Angers, France, on May 4, 1980. After graduating from
high school in July 1998 (Lycée David d’Angers), he pursued a two-year course of higher
education in physics, chemistry, and mathematics (Lycée Bergson, Angers), allowing him to
enter in September 2000 the Graduate School of Chemistry and Physics of Bordeaux (ENSCPB),
France, where he obtained his master’s diploma. In July 2002, he moved to the University of
Florida in Gainesville for his doctoral studies to investigate the air/water interfacial self-
assembly of amphiphilic block copolymers and to synthesize nanoparticles for drug
detoxification applications with Prof. Randolph S. Duran.